Computer Assistive Technologies for Physically and Cognitively Challenged Users 9815079166, 9789815079166

Table of contents :
Cover
Title
Copyright
End User License Agreement
Contents
Preface
List of Contributors
Overview, Category and Ontology of Assistive Devices
Arun Kumar G. Hiremath1,* and Nirmala C.R.1
INTRODUCTION
Scope of the Assistive Technology
Smart Self-management as a Means to Empower with Assistive Technology
Who Adopts Assistive Technology?
The Emergence of Assistive Technology
Professional Practice in Assistive Technology
The Features of Assistive Technology
Categories
No-Technology Devices
Low-Technology Devices
Mid and High Technology Devices
Design Considerations for AT
Evaluation of Functional Capabilities of Assistive Devices
Possible Outcomes with AT
Feature Matching
Ontology of Assistive Devices
General Purpose Assistive Technologies
Performance Areas
Assistive Technology for Manipulation and Control of the Environment
Issues Associated with Assistive Technology Practice
Attempts to Maximize the Accessibility and Affordability of Assistive Technology
Research Trends and Future Research Directions
CONCLUSION
CONSENT FOR PUBLICATION
CONFLICT OF INTEREST
ACKNOWLEDGMENT
REFERENCES
Accessibility of Software/Hardware
Meenu Chandel1,* and Manu Sood1
BACKGROUND
INTRODUCTION
ACCESSIBILITY FOR DIFFERENT CATEGORIES OF PWDS
Visually Impaired Individuals
Physically Challenged Individuals
Deaf and/or Hearing Impaired Individuals
HARDWARE AND SOFTWARE ACCESSIBILITY FOR PWDS
Hardware Options
Software Options
ASSISTIVE TECHNOLOGY
DISABILITIES AND WEB ACCESSIBILITY
DISABILITIES AND ICT ACCESSIBILITY
Frequency of Using ICT Facilities
Challenges Constraining Access to and Use of ICTs by the PwD
Inadequate Friendliness
Ineffective Training Provisions
Power Supply Outages
Outdated ICT Infrastructure
Shortage of ICTs Experts and Technicians
Internet Connectivity
Results of Shortage of ICT Facilities
RECOMMENDATIONS AND SUGGESTIONS
CONCLUSION
CONSENT FOR PUBLICATION
CONFLICT OF INTEREST
ACKNOWLEDGEMENT
REFRENCES
Computer Vision-Based Assistive Technology forBlind and Visually Impaired People: A DeepLearning Approach
Assistive Technology for Home Comfort and Care
Annu Rani1,*, Vishal Goyal1 and Lalit Goyal2
INTRODUCTION
DISABILITY
Types of Disabilities
Blindness
Low Vision
Hearing Disability
Dwarfism
Intellectual Disability
Autism Spectrum Disorder (ASD)
Mental Illness
Locomotor Disability
Leprosy Cured Persons
Muscular Dystrophy (MD)
Chronic Neurological Conditions
Specific Learning Disability
Multiple Sclerosis(MS)
Speech and Language Disability
Thalassemia
Hemophilia
Sickle Cell Disease
Multiple Disabilities, including Deaf-Blindness
Acid Attack
Parkinson’s disease (PD)
Cerebral Palsy (CP)
COMMON BARRIERS FACED BY PEOPLE WITH DISABILITIES
Communication Problem
Physical obstacles
Social Obstacles
Attitudinal barriers
Transportation obstacles
PRINCIPLES FOR PROVIDING ASSISTIVE DEVICES
Availability
Accessibility
Affordability
Adaptability
Acceptability
Quality
ASSISTIVE TECHNOLOGIES FOR HOME RELAXATION AND CARE FOR DISABLED PEOPLE
Mobility aids
Listening and Hearing Aids
Cognitive Devices
Comforting Aids
Limit Motor Skills Aids
Vision Aids
Home Security and Safety
Daily Living Aids
Computer Access Aids
MOBILE APPS FOR ALL DISABILITIES
BENEFITS OF ASSISTIVE TECHNOLOGY DEVICES IN INDIVIDUAL'S LIFE
CONCLUSION
CONSENT OF PUBLICATION
CONFLICT OF INTEREST
ACKNOWLEDGEMENT
REFERENCES
Technologies for Hearing Impaired People UsingIndian Sign Language Synthetic Animations
Augmentative and Alternative Communication/ Hearing Impairments
Jestin Joy1,*, Kannan Balakrishnan2 and M Sreeraj3
INTRODUCTION
BACKGROUND
Sign Language Recognition
Sensor-based System
Vision-based Systems
Challenges and motivation of Sign Language Recognition
Commonly used Sensors
Different Recognition Models
Sign Language Generation
Data Science based AAC Solutions
CONCLUSION AND FUTURE DIRECTIONS
CONSENT FOR PUBLICATION
CONFLICT OF INTEREST
ACKNOWLEDGEMENT
REFERENCES
Hardware and Software-based Accessibility Innovations to Help Physically Disabled User
Bhagvan Kommadi1,*
INTRODUCTION
ACCESSIBILITY FOR DIFFERENT DISABILITIES
CRITICAL ELEMENTS - ACCESSIBILITY ECOSYSTEM
Accessibility Device and Access Options
Vision and Speech Accessibility Options
Speech and Interaction Options
Media and Learning Options
DESIGNING FOR ACCESSIBILITY
Web Accessibility Improvements
BEST PRACTICES
DIGITAL ACCESSIBILITY
ACCESSIBILITY PROJECT LIFECYCLE
PLANNING FOR ACCESSIBILITY
ACCESSIBILITY PLATFORM
Disability - Assistive Technology
Research GAPS
CONCLUSION
CONSENT FOR PUBLICATION
CONFLICT OF INTEREST
ACKNOWLEDGEMENT
REFERENCES
Elderly and Visually Impaired People Mobility inHome Environment Using Adhesive TactileWalking Surface Indicators
Assistive Technology Trends, Challenges and Future Directions
Nancy Jasmine Goldena1,* and Thangapriya1
WHERE ARE WE NOW WITH ASSISTIVE TECHNOLOGY?
The Evolution of AT
Foundation Period (1800 – 1900)
Establishment Period (1900-1972)
Empowerment Period (1972-2010)
Technologically Sophisticated Period(2011-present)
Legal Mandates
IMPORTANCE OF ASSISTIVE TECHNOLOGY
Cognitive Disability
Motor Disability
Visual Disability
Auditory Disability
APPROACHES AND CRITICISMS IN THE CURRENT STUDY OF ASSISTIVE TECHNOLOGY
Approaches of AT
AT for Cognitive Disability
AT for Motor Disability
AT for Visual Disability
AT for Auditory Disability
Criticisms in Implementing AT
LIMITATIONS AND CHALLENGES IN ASSISTIVE TECHNOLOGY
Lack of Awareness
Lack of Governance
Lack of Services
Lack of Products
Lack of Inaccessible Environments
Lack of Human Resources
Lack of Finance
Assistive Technology’s Challenges
Challenges in Availability
Challenges in Accessibility
Challenges in Affordability
Challenges in Adaptability
Challenges in Acceptability
Challenges in Quality
Challenges in Research
Challenges in Policy Implementation
Challenges in Multisectoral Action
FUTURE DIRECTIONS IN ASSISTIVE TECHNOLOGY
Cognitive Disability
Motor Disability
Visual Disability
Auditory Disability
The Following are Some of the Most Recent AT Research Openings
CONCLUSION
CONSENT FOR PUBLICATION
CONFLICT OF INTEREST
ACKNOWLEDGEMENT
REFERENCES
Subject Index
Back Cover

Citation preview

Advances in Data ScienceDriven Technologies (Volume 2) Computer Assistive Technologies for Physically and Cognitively Challenged Users Edited by Manoj Kumar M.V.

Department of Information Science and Engineering Nitte Meenakshi Institute of Technology Bengaluru - 64, Affiliated to Visvesvaraya Technological University Belagavi, Karnataka India

Immanuel Azaad Moonesar R.D.

Associate Professor - Health Administration & Policy Mohammed Bin Rashid School of Government Level 7, Convention Tower P.O. Box 72229 Dubai

President, Academy of International Business- Middle East North Africa Chapter Level 7, Convention Tower P.O. Box 72229, Dubai UAE

Ananth Rao

University of Dubai (AACSB & ABET Accredited) Non-Resident Research Fellow (NRF) Mohammed Bin Rashid School of Government (MBRSG) Dubai United Arab Emirates Adviser Justice KS Hegde Institute of Management (JKSHIM) Nitte University India

Pradeep N.

Department of Computer Science and Engineering Bapuji Institute of Engineering and Technology Davanagere, Affiliated to Visvesvaraya Technological University Belagavi, Karnataka India

Annappa

Department of Computer Science and Engineering National Institute of Technology Karnataka Surathkal, PO Srinivasnagar, Mangalore 575 025 India

Sandeep Kautish

Dean-Academics LBEF Campus, Kathmandu (Nepal)

& Vijayakumar Varadarajan

School of Computer Science and Engineering, UNSW, Australia Swiss School of Business and Management, SSBM, Switzerland School of NUOVOS, ADYPU, India

Advances in Data Science-Driven Technologies (Volume 2) Computer Assistive Technologies for Physically and Cognitively Challenged Users Editors: Manoj Kumar M.V, Immanuel Azaad Moonesar R.D., Ananth Rao, Pradeep N, Annappa, Sandeep Kautish, and Vijayakumar Varadarajan ISSN (Online): 2972-3450 ISSN (Print): 2972-3442 ISBN (Online): 978-981-5079-15-9 ISBN (Print): 978-981-5079-16-6 ISBN (Paperback): 978-981-5079-17-3 © 2023, Bentham Books imprint. Published by Bentham Science Publishers Pte. Ltd. Singapore. All Rights Reserved. First published in 2023.

BSP-EB-PRO-9789815079159-TP-217-TC-09-PD-20230322

BENTHAM SCIENCE PUBLISHERS LTD.

End User License Agreement (for non-institutional, personal use) This is an agreement between you and Bentham Science Publishers Ltd. Please read this License Agreement carefully before using the ebook/echapter/ejournal (“Work”). Your use of the Work constitutes your agreement to the terms and conditions set forth in this License Agreement. If you do not agree to these terms and conditions then you should not use the Work. Bentham Science Publishers agrees to grant you a non-exclusive, non-transferable limited license to use the Work subject to and in accordance with the following terms and conditions. This License Agreement is for non-library, personal use only. For a library / institutional / multi user license in respect of the Work, please contact: [email protected].

Usage Rules: 1. All rights reserved: The Work is the subject of copyright and Bentham Science Publishers either owns the Work (and the copyright in it) or is licensed to distribute the Work. You shall not copy, reproduce, modify, remove, delete, augment, add to, publish, transmit, sell, resell, create derivative works from, or in any way exploit the Work or make the Work available for others to do any of the same, in any form or by any means, in whole or in part, in each case without the prior written permission of Bentham Science Publishers, unless stated otherwise in this License Agreement. 2. You may download a copy of the Work on one occasion to one personal computer (including tablet, laptop, desktop, or other such devices). You may make one back-up copy of the Work to avoid losing it. 3. The unauthorised use or distribution of copyrighted or other proprietary content is illegal and could subject you to liability for substantial money damages. You will be liable for any damage resulting from your misuse of the Work or any violation of this License Agreement, including any infringement by you of copyrights or proprietary rights.

Disclaimer: Bentham Science Publishers does not guarantee that the information in the Work is error-free, or warrant that it will meet your requirements or that access to the Work will be uninterrupted or error-free. The Work is provided "as is" without warranty of any kind, either express or implied or statutory, including, without limitation, implied warranties of merchantability and fitness for a particular purpose. The entire risk as to the results and performance of the Work is assumed by you. No responsibility is assumed by Bentham Science Publishers, its staff, editors and/or authors for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products instruction, advertisements or ideas contained in the Work.

Limitation of Liability: In no event will Bentham Science Publishers, its staff, editors and/or authors, be liable for any damages, including, without limitation, special, incidental and/or consequential damages and/or damages for lost data and/or profits arising out of (whether directly or indirectly) the use or inability to use the Work. The entire liability of Bentham Science Publishers shall be limited to the amount actually paid by you for the Work.

General: 1. Any dispute or claim arising out of or in connection with this License Agreement or the Work (including non-contractual disputes or claims) will be governed by and construed in accordance with the laws of Singapore. Each party agrees that the courts of the state of Singapore shall have exclusive jurisdiction to settle any dispute or claim arising out of or in connection with this License Agreement or the Work (including non-contractual disputes or claims). 2. Your rights under this License Agreement will automatically terminate without notice and without the

need for a court order if at any point you breach any terms of this License Agreement. In no event will any delay or failure by Bentham Science Publishers in enforcing your compliance with this License Agreement constitute a waiver of any of its rights. 3. You acknowledge that you have read this License Agreement, and agree to be bound by its terms and conditions. To the extent that any other terms and conditions presented on any website of Bentham Science Publishers conflict with, or are inconsistent with, the terms and conditions set out in this License Agreement, you acknowledge that the terms and conditions set out in this License Agreement shall prevail. Bentham Science Publishers Pte. Ltd. 80 Robinson Road #02-00 Singapore 068898 Singapore Email: [email protected]

BSP-EB-PRO-9789815079159-TP-217-TC-09-PD-20230322

CONTENTS PREFACE ................................................................................................................................................ i LIST OF CONTRIBUTORS .................................................................................................................. iii CHAPTER 1 OVERVIEW, CATEGORY AND ONTOLOGY OF ASSISTIVE DEVICES ........ Arun Kumar G. Hiremath and Nirmala C.R. INTRODUCTION .......................................................................................................................... Scope of the Assistive Technology ......................................................................................... Smart Self-management as a Means to Empower with Assistive Technology ...................... Who Adopts Assistive Technology? ....................................................................................... The Emergence of Assistive Technology ............................................................................... Professional Practice in Assistive Technology ....................................................................... The Features of Assistive Technology .................................................................................... Categories ............................................................................................................................... No-Technology Devices ......................................................................................................... Low-Technology Devices ....................................................................................................... Mid and High Technology Devices ........................................................................................ Design Considerations for AT ................................................................................................ Evaluation of Functional Capabilities of Assistive Devices ................................................... Possible Outcomes with AT .................................................................................................... Feature Matching .................................................................................................................... Ontology of Assistive Devices ................................................................................................ General Purpose Assistive Technologies ................................................................................ Performance Areas .................................................................................................................. Assistive Technology for Manipulation and Control of the Environment ............................. Issues Associated with Assistive Technology Practice .......................................................... Attempts to Maximize the Accessibility and Affordability of Assistive Technology ............ Research Trends and Future Research Directions .................................................................. CONCLUSION ............................................................................................................................... CONSENT FOR PUBLICATION ................................................................................................ CONFLICT OF INTEREST ......................................................................................................... ACKNOWLEDGMENT ................................................................................................................ REFERENCES ...............................................................................................................................

1

CHAPTER 2 ACCESSIBILITY OF SOFTWARE/HARDWARE .................................................. Meenu Chandel and Manu Sood BACKGROUND ............................................................................................................................. INTRODUCTION .......................................................................................................................... ACCESSIBILITY FOR DIFFERENT CATEGORIES OF PWDS ........................................... Visually Impaired Individuals ................................................................................................. Physically Challenged Individuals .......................................................................................... Deaf and/or Hearing Impaired Individuals ............................................................................. HARDWARE AND SOFTWARE ACCESSIBILITY FOR PWDS .......................................... Hardware Options ................................................................................................................... Software Options .................................................................................................................... ASSISTIVE TECHNOLOGY ....................................................................................................... DISABILITIES AND WEB ACCESSIBILITY ........................................................................... DISABILITIES AND ICT ACCESSIBILITY ............................................................................. Frequency of Using ICT Facilities .......................................................................................... Challenges Constraining Access to and Use of ICTs by the PwD ......................................... Inadequate Friendliness ................................................................................................

26

1 2 3 3 4 4 5 7 8 9 10 11 11 12 13 13 16 16 18 20 21 22 22 23 23 23 23

26 28 29 29 30 30 30 30 31 32 36 38 41 41 41

Ineffective Training Provisions ..................................................................................... Power Supply Outages .................................................................................................. Outdated ICT Infrastructure ......................................................................................... Shortage of ICTs Experts and Technicians ................................................................... Internet Connectivity ..................................................................................................... Results of Shortage of ICT Facilities ...................................................................................... RECOMMENDATIONS AND SUGGESTIONS ........................................................................ CONCLUSION ............................................................................................................................... CONSENT FOR PUBLICATION ................................................................................................ CONFLICT OF INTEREST ......................................................................................................... ACKNOWLEDGEMENT ............................................................................................................. REFRENCES .................................................................................................................................. CHAPTER 3 COMPUTER VISION-BASED ASSISTIVE TECHNOLOGY FOR BLIND AND VISUALLY IMPAIRED PEOPLE: A DEEP LEARNING APPROACH ......................................... Roopa G.M., Chetana Prakash and Pradeep N. INTRODUCTION .......................................................................................................................... THE GLOBAL ASSISTIVE TECHNOLOGY COMMUNITY AND ITS IMPACTS ON PEOPLE WITH DISABILITIES .................................................................................................. PRESENT-DAY SCENARIO ........................................................................................................ GENERAL DESIGN IDEAS AND THE USABILITY OF DAILY ITEMS ............................. EVOLUTION OF ASSISTIVE TECHNOLOGIES .................................................................... ASSISTIVE TECHNOLOGIES: FUNCTIONAL FRAMEWORK .......................................... Hard-Soft Technologies .......................................................................................................... OBJECT RECOGNITION ............................................................................................................ BACKGROUND THEORY ........................................................................................................... Object Detection Algorithms .................................................................................................. SIFT (Scale Invariant Feature Transform) Algorithm .................................................. SURF (Speeded Up Robust Features) ........................................................................... OCR(Optical-Character-Recognition) .......................................................................... YOLO (You Only Look Once) ....................................................................................... R-CNN ........................................................................................................................... Gaps Identified ........................................................................................................................ Existing Assistance solutions for Blind People ...................................................................... PRIMARY OBJECTIVE OF COMPUTER VISION ................................................................. METHODOLOGY PROPOSED .................................................................................................. YOLOV3 ARCHITECTURE ........................................................................................................ EXPERIMENTAL SETUP ............................................................................................................ RESULTS AND DISCUSSION ..................................................................................................... System Work-Flow for Object Detection ............................................................................... SMART READING SYSTEM FOR VISUALLY IMPAIRED PEOPLE USING TESSERACT ................................................................................................................................... FLOW PROCESS OF TESSERACT ........................................................................................... FUTURE RESEARCH DIRECTIONS ........................................................................................ CONCLUSION ............................................................................................................................... CONSENT OF PUBLICATION ................................................................................................... CONFLICT OF INTEREST ......................................................................................................... ACKNOWLEDGMENT ................................................................................................................ REFERENCES ...............................................................................................................................

42 42 42 43 43 43 44 44 45 45 45 45 48 49 51 52 52 53 54 54 56 57 57 57 57 58 58 59 59 59 61 61 62 63 65 66 67 67 69 69 70 70 70 70

CHAPTER 4 ASSISTIVE TECHNOLOGY FOR HOME COMFORT AND CARE .................... 73 Annu Rani, Vishal Goyal and Lalit Goyal

INTRODUCTION .......................................................................................................................... DISABILITY ................................................................................................................................... Types of Disabilities ............................................................................................................... Blindness ....................................................................................................................... Low Vision ..................................................................................................................... Hearing Disability ......................................................................................................... Dwarfism ....................................................................................................................... Intellectual Disability .................................................................................................... Autism Spectrum Disorder (ASD) ................................................................................. Mental Illness ................................................................................................................ Locomotor Disability .................................................................................................... Leprosy Cured Persons ................................................................................................. Muscular Dystrophy (MD) ............................................................................................ Chronic Neurological Conditions ................................................................................. Specific Learning Disability .......................................................................................... Multiple Sclerosis(MS) .................................................................................................. Speech and Language Disability ................................................................................... Thalassemia ................................................................................................................... Hemophilia .................................................................................................................... Sickle Cell Disease ........................................................................................................ Multiple Disabilities, including Deaf-Blindness ........................................................... Acid Attack .................................................................................................................... Parkinson’s disease (PD) .............................................................................................. Cerebral Palsy (CP) ...................................................................................................... COMMON BARRIERS FACED BY PEOPLE WITH DISABILITIES .................................. Communication Problem ........................................................................................................ Physical obstacles ................................................................................................................... Social Obstacles ...................................................................................................................... Attitudinal barriers .................................................................................................................. Transportation obstacles ......................................................................................................... PRINCIPLES FOR PROVIDING ASSISTIVE DEVICES ........................................................ Availability ............................................................................................................................. Accessibility ............................................................................................................................ Affordability ........................................................................................................................... Adaptability ............................................................................................................................. Acceptability ........................................................................................................................... Quality ..................................................................................................................................... ASSISTIVE TECHNOLOGIES FOR HOME RELAXATION AND CARE FOR DISABLED PEOPLE ..................................................................................................................... Mobility aids ........................................................................................................................... Listening and Hearing Aids .................................................................................................... Cognitive Devices ................................................................................................................... Comforting Aids ..................................................................................................................... Limit Motor Skills Aids .......................................................................................................... Vision Aids ............................................................................................................................. Home Security and Safety ...................................................................................................... Daily Living Aids ................................................................................................................... Computer Access Aids ............................................................................................................ MOBILE APPS FOR ALL DISABILITIES ................................................................................ BENEFITS OF ASSISTIVE TECHNOLOGY DEVICES IN INDIVIDUAL'S LIFE ............ CONCLUSION ...............................................................................................................................

73 74 75 75 75 75 75 75 76 76 76 76 76 76 77 77 77 77 77 77 78 78 78 78 79 79 79 80 80 80 81 81 81 81 82 82 82 82 82 84 85 85 85 86 88 88 89 90 94 94

CONSENT OF PUBLICATION ................................................................................................... CONFLICT OF INTEREST ......................................................................................................... ACKNOWLEDGEMENT ............................................................................................................. REFERENCES ............................................................................................................................... CHAPTER 5 TECHNOLOGIES FOR HEARING IMPAIRED PEOPLE USING INDIAN SIGN LANGUAGE SYNTHETIC ANIMATIONS ........................................................................................ Rakesh Kumar, Lalit Goyal and Vishal Goyal INTRODUCTION .......................................................................................................................... FACTS ABOUT INDIAN SIGN LANGUAGE ........................................................................... COMMUNICATION BETWEEN DEAF AND HEARING COMMUNITIES ........................ ENGLISH TEXT TO INDIAN SIGN LANGUAGE TRANSLATION SYSTEM ................... English-ISL Lexicon ............................................................................................................... Text Parser Module to Parse English Sentences ..................................................................... Grammatical Rules for Transformation of English to ISL Sentence ...................................... Eliminator Module for Removal of Undesired Words ........................................................... Lemmatization and Synonym Replacement ........................................................................... Sign Animation using Avatar .................................................................................................. ANNOUNCEMENTS SYSTEM FOR RAILWAY STATIONS ................................................ ANNOUNCEMENTS SYSTEM FOR AIRPORTS .................................................................... ANNOUNCEMENTS SYSTEM FOR BUS STANDS ................................................................. CONCLUSION AND FUTURE WORK ...................................................................................... CONSENT FOR PUBLICATION ................................................................................................ CONFLICT OF INTEREST ......................................................................................................... ACKNOWLEDGEMENT ............................................................................................................. REFERENCES ............................................................................................................................... CHAPTER 6 AUGMENTATIVE AND ALTERNATIVE COMMUNICATION/ HEARING IMPAIRMENTS ...................................................................................................................................... Jestin Joy, Kannan Balakrishnan and M Sreeraj INTRODUCTION .......................................................................................................................... BACKGROUND ............................................................................................................................. Sign Language Recognition .................................................................................................... Sensor-based System .............................................................................................................. Vision-based Systems ............................................................................................................. Challenges and motivation of Sign Language Recognition .................................................... Commonly used Sensors ......................................................................................................... Different Recognition Models ................................................................................................ Sign Language Generation ...................................................................................................... Data Science based AAC Solutions ........................................................................................ CONCLUSION AND FUTURE DIRECTIONS .......................................................................... CONSENT FOR PUBLICATION ................................................................................................ CONFLICT OF INTEREST ......................................................................................................... ACKNOWLEDGEMENT ............................................................................................................. REFERENCES ............................................................................................................................... CHAPTER 7 HARDWARE AND SOFTWARE-BASED ACCESSIBILITY INNOVATIONS TO HELP PHYSICALLY DISABLED USER ............................................................................................ Bhagvan Kommadi INTRODUCTION .......................................................................................................................... ACCESSIBILITY FOR DIFFERENT DISABILITIES ............................................................. CRITICAL ELEMENTS - ACCESSIBILITY ECOSYSTEM ..................................................

95 95 95 95 98 99 101 102 103 105 107 107 109 109 110 111 112 113 114 115 115 115 116 117 117 119 120 121 122 123 124 125 127 129 130 130 130 130 131 135 135 136 138

Accessibility Device and Access Options ............................................................................... Vision and Speech Accessibility Options ............................................................................... Speech and Interaction Options .............................................................................................. Media and Learning Options .................................................................................................. DESIGNING FOR ACCESSIBILITY .......................................................................................... Web Accessibility Improvements ........................................................................................... BEST PRACTICES ........................................................................................................................ DIGITAL ACCESSIBILITY ......................................................................................................... ACCESSIBILITY PROJECT LIFECYCLE ............................................................................... PLANNING FOR ACCESSIBILITY ........................................................................................... ACCESSIBILITY PLATFORM ................................................................................................... Disability - Assistive Technology ........................................................................................... Research GAPS ....................................................................................................................... CONCLUSION ............................................................................................................................... CONSENT FOR PUBLICATION ................................................................................................ CONFLICT OF INTEREST ......................................................................................................... ACKNOWLEDGEMENT ............................................................................................................. REFERENCES ............................................................................................................................... CHAPTER 8 ELDERLY AND VISUALLY IMPAIRED PEOPLE MOBILITY IN HOME ENVIRONMENT USING ADHESIVE TACTILE WALKING SURFACE INDICATORS .......... Vijaya Prakash R. and Srinath Taduri INTRODUCTION .......................................................................................................................... RELATED WORK ......................................................................................................................... TACTILE DESIGN METHODOLOGY ...................................................................................... Target Users ............................................................................................................................ Tactile Design ......................................................................................................................... Color Experimentation .................................................................................................. Foot Sensitivity Test ...................................................................................................... Surface Texture Test ...................................................................................................... Tactile Test .............................................................................................................................. Tile Experiments ..................................................................................................................... RESULTS AND DISCUSSION ..................................................................................................... CONCLUSION ............................................................................................................................... CONSENT TO PUBLISH .............................................................................................................. CONFLICT OF INTEREST ......................................................................................................... ACKNOWLEDGEMENTS ........................................................................................................... REFERENCES ............................................................................................................................... CHAPTER 9 ASSISTIVE TECHNOLOGY TRENDS, CHALLENGES AND FUTURE DIRECTIONS .......................................................................................................................................... Nancy Jasmine Goldena and Thangapriya WHERE ARE WE NOW WITH ASSISTIVE TECHNOLOGY? ............................................. The Evolution of AT ............................................................................................................... Foundation Period (1800 – 1900) ........................................................................................... Establishment Period (1900-1972) ......................................................................................... Empowerment Period (1972-2010) ......................................................................................... Technologically Sophisticated Period(2011-present) ............................................................. Legal Mandates ....................................................................................................................... IMPORTANCE OF ASSISTIVE TECHNOLOGY .................................................................... Cognitive Disability ................................................................................................................ Motor Disability ......................................................................................................................

139 140 141 142 143 144 147 151 153 156 158 159 160 161 161 161 161 161 164 164 166 169 169 170 170 170 171 171 172 175 179 179 180 180 180 183 184 184 184 184 185 186 187 188 190 191

Visual Disability ..................................................................................................................... Auditory Disability ................................................................................................................. APPROACHES AND CRITICISMS IN THE CURRENT STUDY OF ASSISTIVE TECHNOLOGY ............................................................................................................................. Approaches of AT ................................................................................................................... AT for Cognitive Disability .................................................................................................... AT for Motor Disability .......................................................................................................... AT for Visual Disability ......................................................................................................... AT for Auditory Disability ..................................................................................................... Criticisms in Implementing AT .............................................................................................. LIMITATIONS AND CHALLENGES IN ASSISTIVE TECHNOLOGY ............................... Lack of Awareness .................................................................................................................. Lack of Governance ................................................................................................................ Lack of Services ...................................................................................................................... Lack of Products ..................................................................................................................... Lack of Inaccessible Environments ........................................................................................ Lack of Human Resources ...................................................................................................... Lack of Finance ....................................................................................................................... Assistive Technology’s Challenges ........................................................................................ Challenges in Availability ....................................................................................................... Challenges in Accessibility ..................................................................................................... Challenges in Affordability ..................................................................................................... Challenges in Adaptability ...................................................................................................... Challenges in Acceptability .................................................................................................... Challenges in Quality .............................................................................................................. Challenges in Research ........................................................................................................... Challenges in Policy Implementation ..................................................................................... Challenges in Multisectoral Action ........................................................................................ FUTURE DIRECTIONS IN ASSISTIVE TECHNOLOGY ...................................................... Cognitive Disability ................................................................................................................ Motor Disability ...................................................................................................................... Visual Disability ..................................................................................................................... Auditory Disability ................................................................................................................. The Following are Some of the Most Recent AT Research Openings ................................... CONCLUSION ............................................................................................................................... CONSENT FOR PUBLICATION ................................................................................................ CONFLICT OF INTEREST ......................................................................................................... ACKNOWLEDGEMENT ............................................................................................................. REFERENCES ...............................................................................................................................

192 192 193 193 193 195 196 197 198 199 200 200 200 200 201 201 201 202 202 202 203 203 203 203 203 204 204 205 205 205 206 206 206 208 208 209 209 209

SUBJECT INDEX .................................................................................................................................... 211

i

PREFACE This book aims to collate the methods and literature related to techniques that will aid the life of cognitively challenged individuals. A cognitive impairment (also known as an intellectual disability) is a term used when a person has certain mental functioning limitations and skills, such as communication, self-help, and social skills. The content presented in this book discusses the range of methods/techniques that will improve the life of a person with cognition problems. The range of topics like the ontology of cognitive devices, accessibility hardware and software, assistive technologies for Vision impairment, hearing impairment and communication impairment has been detailed extensively. This edited book also sheds light on upcoming trends, challenges, and future research directions in assistive technologies for cognitively challenged users. We editors believe this book will help researchers, students, academicians and medical practitioners know and adopt state-of-the-art technologies in cognitive disability. We extend our heartfelt thanks to our reviewers, who have extended their support despite their busy schedules. A special thanks to all our authors for submitting the work. Our sincere thanks to Bentham Science publishers for accepting our proposal for editing this book and supporting us extensively during the editing process. Our thanks to one and all who have directly or indirectly rendered support for completing this edited book. We believe the efforts we rendered for editing the book are worthwhile only if this book is of any use to the ordinary end-users of our society. This satisfaction will fuel us to come up with more edited books that will be useful for society at large.

Manoj Kumar M.V Department of Information Science and Engineering Nitte Meenakshi Institute of Technology Bengaluru - 64, Affiliated to Visvesvaraya Technological University Belagavi, Karnataka India Immanuel Azaad Moonesar R.D. Associate Professor - Health Administration & Policy Mohammed Bin Rashid School of Government Level 7, Convention Tower P.O. Box 72229 Dubai President, Academy of International Business- Middle East North Africa Chapter Level 7, Convention Tower P.O. Box 72229, Dubai UAE Ananth Rao 1.University of Dubai (AACSB & ABET Accredited) 2. Non-Resident Research Fellow (NRF) Mohammed Bin Rashid School of Government (MBRSG) Dubai

ii

United Arab Emirates 3. Adviser Justice KS Hegde Institute of Management (JKSHIM) Nitte University India Pradeep N Department of Computer Science and Engineering Bapuji Institute of Engineering and Technology Davanagere, Affiliated to Visvesvaraya Technological University Belagavi, Karnataka India Annappa Department of Computer Science and Engineering National Institute of Technology Karnataka Surathkal, PO Srinivasnagar, Mangalore 575 025 India Sandeep Kautish Dean-Academics LBEF Campus, Kathmandu (Nepal) & Vijayakumar Varadarajan School of Computer Science and Engineering The University of New South Wales Sydney Australia

iii

List of Contributors Annu Rani

Department of Computer Science, Punjabi University, Patiala, India

Arun Kumar G. Hiremath

Department of Computer Science and Engineering, Bapuji Institute of Engineering and Technology, Davangere – 577004, Visvesvaraya Technological University, Belagavi – 590 018

Bhagvan Kommadi

Director of Product Engineering, Value Momentum, Hyderabad, India

Chetana Prakash

Department of Computer Science and Engineering, Bapuji Institute of Engineering and Technology, Davangere – 577004, Visvesvaraya Technological University, Belagavi – 590 018

Jestin Joy

Department of Computer Applications, St. George’s College, Aruvithura, Kerala, India

Kannan Balakrishnan

Department of Computer Applications, CUSAT, Kerala, India

Lalit Goyal

Department of Computer Science, DAV College, Jalandhar (Punjab, India

M Sreeraj

Sree Ayyappa College, Eramallikkara, Alappuzha, Kerala, India

Manu Sood

Department of Computer Science, Himachal Pradesh University, Shimla, India

Meenu Chandel

Department of Computer Science, Himachal Pradesh University, Shimla, India

Nancy Jasmine Goldena

Department of Computer Applications and Research Centre, , Sarah Tucker College(Autonomous), Tirunelveli, Tamilnadu, India

Nirmala C.R.

Department of Computer Science and Engineering, Bapuji Institute of Engineering and Technology, Davangere – 577004, Visvesvaraya Technological University, Belagavi – 590 018

Pradeep N.

Department of Computer Science and Engineering, Bapuji Institute of Engineering and Technology, Davangere – 577004, Visvesvaraya Technological University, Belagavi – 590 018

Rakesh Kumar

Department of Computer Science,, University College Miranpur, Patiala, India

Roopa G.M.

Department of Computer Science and Engineering, Bapuji Institute of Engineering and Technology, Davangere – 577004, Visvesvaraya Technological University, Belagavi – 590 018

Srinath Taduri

Department of Computer Science and Engineering, SR Engineering College, Waranga, India

Thangapriya

Department of Computer Applications and Research Centre, , Sarah Tucker College(Autonomous), Tirunelveli, Tamilnadu, India

Vijaya Prakash R.

Department of Computer Science and Engineering, SR Engineering College, Waranga, India

Vishal Goyal

Department of Computer Science, Punjabi University, Patiala, India

Advances in Data Science-Driven Technologies, 2023, 1-25

1

CHAPTER 1

Overview, Category and Ontology of Assistive Devices Arun Kumar G. Hiremath1,* and Nirmala C.R.1 Department of Computer Science and Engineering, Bapuji Institute of Engineering and Technology, Davangere – 577004, Visvesvaraya Technological University, Belagavi – 590 018 1

Abstract: The majority of physically challenged and elderly people demand a lot of care when it comes to assistive technologies that can provide tailored services to their needs. The primary concern of advancement in Assistive technology is to address a wide variety of disabilities and intellectual impairments for societal benefits by reducing welfare costs and allowing for an efficient workforce. To better respond to changes brought on by modernity, it is necessary to understand how assistive technology interacts in that group. The broad range of assistive devices in the continuum of assistive technology can help people with various impairments. Based on the underlying technology, the Categorization of assistive devices has important implications for clinical usage when examined through the perspective of social phenomenon. In the realm of Assistive Technology, a consistent focus on the relationship between the individual and the supported activity within certain contexts is essential. Assistive technology can be viewed from the perspective of various performance areas. The Ontology-based Assistive Devices that are among the finest within common, everyday contexts for more relevant applications are interesting. This chapter explores all those essential elementary and general considerations of assistive devices that form the bases of Assistive technology and brings out the categories of assistive devices and the various application domains where assistive devices can be served as a derivative of a particular ontology. The chapter focuses on the various performance areas by addressing the issues associated with Assistive technology Practice.

Keywords: Accessibility, Assistive technology, Cognitive impairment, Information and Communication Technology (ICT), Ontology, Self-management. INTRODUCTION The population of aging adults is expected to reach more than two billion by 2050. In a society where the life expectancy and increasing need for assistance are Corresponding author Arun Kumar G Hiremath: Department of Computer Science and Engineering, Bapuji Institute of Engineering and Technology, Davangere- 577004; E-mail:[email protected]

*

Manoj Kumar M V, Immanuel Azaad Moonesar R.D., Ananth Rao, Pradeep N, Annappa, Sandeep Kautish and Vijayakumar Varadarajan (Eds.) All rights reserved-© 2023 Bentham Science Publishers

2 Advances in Data Science-Driven Technologies

Hiremath and C.R.

advancing, it is becoming more likely that elderly people will need the technology to accomplish critical and necessary tasks. Assistive technology is the most prominent and prime solution that exemplifies how technologies can be used to meet the requirements of the elderly. People with disabilities, those who live longer, those with non-communicable disorders, and those needing recovery are all potential beneficiaries of assistive technologies, which help them live independently and enable them to maintain their dignity. In a broader sense, assistive technology is needed for all people with cognitive/physical disabilities, mental health disorders, progressive functional impairment, non-communicable diseases, etc. Assistive technology aims to ensure that any artificial aid a patient takes, requires no external dependence. For the moral well-being of the patients, it is important that they feel independent and can manage the majority of their tasks on their own. Assistive Technology (AT) is either an element or a piece of equipment used to enhance, preserve, or expand the associated support of an impaired individual's life. Reasonable assistive technology may also help individuals accommodate a disability, at least partially. Traditionally, the word “assistive technology” has been used to refer to computer software and hardware, as well as digital equipment. Assistive Technology is a broad term that refers to a range of low- to high-tech devices whose major intention is to enhance a person's individual functioning and mobility in order to maximize involvement and greatly improve quality of life. Mobility aids, such as prosthetic devices and orthotic devices, cognitive aids, such as electronic or electrical assistive devices, and high-performance mobile devices that enable people with disabilities to participate in sports and be physically active are some of the examples. They can also help avoid impairments and secondary health problems by encouraging independence and autonomy in the person and those around them. Scope of the Assistive Technology Assistive technology offers opportunities for every individual with a disability by providing the most appropriate technologies and removing environmental barriers to functioning. Computers are the entities most widely associated with Assistive Technology. However, a broad spectrum of Assistive Technology ranges from mainstream gadgets to exoskeletons and robotics, sophisticated automated systems, intelligent houses, etc. The technology support includes ergonomics and telerehabilitation with the aid of environmental accommodations and service delivery systems.

Assistive Devices

Advances in Data Science-Driven Technologies 3

People with learning difficulties are increasingly turning to assistive technologies for help. Nevertheless, general computer use is a relatively widespread phenomenon, as seen by the availability of computers for a wide range of applications. The potential advancement in the computer environment has changed the nature of technology support. First, in the last decade, technology has emerged as a platform where powerful yet cheaper modern equipment can be afforded. Second, a lot of new technologies have developed. Third, the sophistication of technology has improved significantly, especially in the realm of computer software. Traditional technology has little in common with modern technology, which features realistic sound, spectacular images, and on-screen videos. According to the current consensus, computer technology and other innovations have a great deal of potential for improving the capacities of children, teens, and adults with learning difficulties. Smart Self-management as a Means to Empower with Assistive Technology Equal opportunities are everyone's rights, but people with disabilities are often ostracised, marginalized, and driven into poverty, which intensifies the impact of psychological distress on a person's social environment and makes it critical to provide helpful services to individuals with a diverse range of impairments. Selfmanagement skills refer to the capabilities to govern one's beliefs and actions. A self-motivated, physically challenged individual can strengthen confidence to manage potential tasks with significant and precise technology-driven assistance. An assistive device-based task accomplishment paradigm can enhance an individual's self-management ability by solving ongoing issues and assignments. There are two main goals of assistive technology. First, it can enhance an individual's strengths so that personal abilities can compensate for any impairments. Second, technologies can provide an alternative means of executing a task, allowing for compensation or eliminating limitations. Who Adopts Assistive Technology? The typical assistive technology user has an impairment that necessitates using a compensatory solution in an attempt to gain more independence. The user's ability or disability can vary. It might range from someone who has a spinal cord injury and can only move their head to someone who suffers from carpal tunnel syndrome and has pain when opening their mouth. Technology can be beneficial to both adults and children. Individuals with a short illness, a long-standing impairment, or a neurologic condition in which the individual's functional abilities will continue to deteriorate are almost all illustrations of AT users.

4 Advances in Data Science-Driven Technologies

Hiremath and C.R.

The Emergence of Assistive Technology The Foundation Period - AT started when the population of people with disabilities, injuries, and troops who survived the wars increased in the early 1990s. As disability disciplines were established during this foundation period to motivate independence and productivity, social perception towards individuals with disabilities changed positively. As more people began to live independently, the use of AT increased. Different acts were passed to help and give rights to individuals with disabilities. Many things that were invented in the past are still being used today. The ear trumpet, which emerged during the 17th century, remained widely accepted in various sizes and shapes. In 1800, various educational fraternities adopted a systematic approach to the blind invented by Louis Braille. Most sign languages are legally recognized. With the aid of a microphone and telephony, hearing aids were developed in the 19th century. While wheelchairs have been used for generations, the first lightweight, foldable wheelchair, which employed an X-joint to allow it to be flattened, was designed in the 19th century. Empowerment Period - The empowerment period is when individuals with disabilities are given the right to pursue their life goals. Many legislations passed to improve the rights of individuals with disabilities, such as the rehabilitation act 1973 and the individuals with disability education act 1997. Most notable AT technologies, such as Closed Caption Television (CCTV), talking calculators, and the very first prototype of a voice synthesizer, were developed during the empowering period. Most ATs were invented during this period to improve independence and achieve goals for individuals with visual and hearing impairments [1]. People realized the “desire to accomplish” during this empowerment period. Many innovative thinkers and scientists have looked for the technical possibilities to indulge the lives of people with physical disabilities and other associated problems. Every technological advancement, from modest prosthetics in the middle ages to complex electronic systems, aims to improve the quality of a person's life as much as possible. Professional Practice in Assistive Technology Assistive technology encompasses selecting, locating, and using assistive and rehabilitative devices for persons with impairments. It's important to realize that this sector has much fragmentary development despite its enormous potential and

Assistive Devices

Advances in Data Science-Driven Technologies 5

reach. The domain is fiercely competitive, but it needs the right assistive technology solutions to achieve it. However, the following organizations which create and pay for administrative assistance devices and workplace technologies primarily employ them: Private-Health-Insurance- When a medical practitioner suggests assistive technology as a vital rehabilitative aid, healthcare entities can employ various assistive technologies that make it easier for nurses to provide quality care to their patients. Business Employers- In a business organization, Assistive technology can be used to provide fair compensation for employees to complete vital activities assigned to them at work. Job Training Programs and Rehabilitation - They use this technology to assist individuals in finding jobs. With ease in the training process and more skills added with the help of AI and ML solutions in everyday life, it becomes quite easier for new people to join interesting professions. School System- They use it for conventional school educational resources as well as technology that is part of Individualized Education Programs (IEP). AI and ML have transformed the Healthcare, Business, Education, and other sectors and welcome the new technology for efficient services. Over the years, we have transformed how things could be better worked on and what solutions can make it easier for the patients to meet their needs. The Features of Assistive Technology When exploring various impairments that might be addressed with assistive technology, it's crucial to understand how each impairment is actually dealt with and what the response might be to treating such concerns in everyday life. Assistive technology aims to make sure that any artificial aid that a patient takes requires no external dependence. For the moral well-being of the patients, it is important that they feel independent and can manage the majority of their tasks on their own. It is here that Assistive Technology comes into the picture. Numerous IT service companies worldwide are constantly improving Assistive Technology and its associated solutions. Various supplemental technological innovations and strategic initiatives have been devised, all of which are uncomplicated to use, precise, and approachable. Hearing: There is a variety of approaches to present sound visualization for

6 Advances in Data Science-Driven Technologies

Hiremath and C.R.

people with hearing impairment to provide a fresh experience. Users can use options like Mono Audio, Adapt Sound, Flash Notification, Turn Off All Sounds, Sound Balance, and Create Vibration after the hearing aid keeps track of incoming messages and guarantees they won't miss sound notifications. Vision: The device's revolutionary and innovative features allow users to view it with ease and get the most out of it in practically any environment, even if the screen isn't visible. Dexterity: When using several gestures the device requires is difficult, a user can control them with an intuitive interface. A User-friendly and streamlined user interface makes it simple to access the options with Assistant Menu. Interaction: Users frequently make unintentional touches on their smartphone screens; the service allows them to adjust the harsh response from the target area. As a result, the device's touch control is more precise, and the operator may accomplish quite well with the gadget by simply touching it just once. Based on the function being performed by the Assistive Technology within the intervention, the AT devices can be categorized into two broad classes: those that are involved in the action prioritization and those that impact self-awareness. The chapter initially focuses on the categories which include the most commonly used assistive technology devices, such as self-care, communication, and safety devices, independent living aids, medication aids, incontinence supplies/aids, reading, and vision aids, home modifications, vehicle modifications, accessible vans, etc. Under this section, a focus has been made on design considerations, evaluation of functional capabilities, and possible outcomes of assistive devices. In Section 3, the ontology of assistive devices has been explored in which the mechanism of usage of technology concepts and relationships to enable standards for a community of humans with various impairments are discussed. A deal with performance areas has been made in Section 4, which includes Assistive Technology for Cognitive Augmentation, AT for Enabling Mobility and Transportation, Assistive technology for home comfort, Controlling the Environment, AT Consideration for Emergency Situations. Further, Assistive Technology Applications have been discussed in various contexts, such as Workplace, Healthcare Industry, etc. The issues and challenges of professional practice and efforts to make Assistive Technology Practice more affordable and accessible have been reviewed. The objective of the chapter is to emphasize the exploration of assistive devices and fundamental performance areas, where assistive devices can meet the purpose of employment. The general considerations and elementary features have been described to provide an understanding of various assistive devices. The

Assistive Devices

Advances in Data Science-Driven Technologies 7

organization of the chapter is as follows. The initial part deals with the categories of assistive devices based on the underlying technology involvement. Next, as a part of framing the categories, the design considerations, evaluation of functional capabilities, and possible outcomes of Assistive devices are discussed. Further, performance areas have been made focusing on ontology and general-purpose assistive technologies. Various issues associated with assistive technology practice and attempts to maximize the accessibility and affordability of Assistive technology are also discussed. Categories Most assistive devices exhibit a multifunctional nature, which makes it difficult to find a proper categorization framework. However, there are certain ways to categorize Assistive Technology Devices. Based on the characteristics, the devices can be classified into two groups: those that influenced self-awareness and those that required action prioritization [2]. Nevertheless, it is unsurprising that Assistive Devices may fall into more than one category, depending on the individual's needs and how and where the person uses the AT. As a result, grouping Assistive Devices based on the underlying technology, nature of the operation, acquisition ability, and associated cost will be a better practice. As a result, it can be seen that assistive technology can range from no and low-tech to high-tech solutions (Fig. 1).

Fig. (1). Categories of Assistive technology.

8 Advances in Data Science-Driven Technologies

Hiremath and C.R.

The use of Assistive devices moves along the Continuum of Assistive technology from no and low-tech to high-tech, based upon the needs of the individual [3]. And hence, Assistive technology tools fall into one of these categories: Notechnology devices, Low Technology devices, Mid-Technology devices, and High Technology devices. Fig. (2) shows most of the AT devices that are categorized based on the underlying technology support.

Fig. (2). Infographic of Assistive technology category.

No-Technology Devices No-tech AT devices require absolutely no machinery and can be as incredibly easy to make as one desires. These are services that rely on procedures and services already in existence in the environment rather than relying on devices or technology. Simple, non-electronic solutions that provide access and improve functional capacity are referred to as No-tech AT. Devices like modified spoon handles, custom-designed pencil grips, and picture communication displays are

Assistive Devices

Advances in Data Science-Driven Technologies 9

examples of AT solutions that can be made or purchased commercially at a cheap cost. The most common No Tech Solution involves Changing the environment, rules, or practices to reduce obstacles without bringing additional items or equipment. In lieu of overhead lighting, using the lights, modifying seating arrangements to provide a comfortable space for someone to work, using threering binders instead of binder clips to hold papers together, and visible work schedule in the classroom environment, Post-it-Notes during reading instruction and a number line during numeracy instruction are the most common illustrations of employing the No technology devices. Low-Technology Devices Low Tech Assistive Technology involves adaptations of very few or limited electronic components. They are relatively cheaper than electronic and digital tools and are often repurposed tools and items that were not originally intended to be assistive technology. Low-tech tools tend to be more readily attainable and easier for the user to learn. These are usually low-priced/affordable or easy to make by using disposable or inexpensive materials. Low-tech devices can be used and manipulated by the individual independently. Low tech is a word that refers to out-of-date technology that is intended to be as minimal as possible. Low-tech does not necessitate the use of a power source. This is sometimes the simplest and most practical assistive technology solution. An individual can have the best support of high-tech assistive technology if he always has a low-tech “backup”. Technology can malfunction, batteries can drain, and disasters can strike, and in such cases, using their low-tech equipment, people can continue to meet their needs. Low-tech devices may also be all that a user requires. For some people, high-tech is not the best option. In a regular school supply store, there are numerous low-tech reading and writing Assistive Technologies. Individuals with low manual dexterity can utilize writing gadgets like Pencil grips or Oversized Pencils to assist them in acquiring a better grasp on writing instruments. Slant & Clipboards help handle documents in position at an inclination, making it convenient for those individuals who cannot turn their wrists to try writing on a flat plane or anyone who has difficulties holding their paper in position during writing. Note-taking assistance, including such enhanced line papers or signature cards, makes it easier to write on a slip of paper. When used with learning methodologies, reading gadgets such as Highlighters and Sticky Notes provide structured learning help and reminders of essential concepts. Reading Aids makes it easier to read by moving over the pages and only obeying one piece of text at a time. Page Magnifiers enlarge the text on a printed page, making it simpler to read.

10 Advances in Data Science-Driven Technologies

Hiremath and C.R.

Mid and High Technology Devices The term “mid-tech” refers to a relatively new technology platform that combines certain advanced features. Consequently, it is widely considered that both mid and high technologies are always superior to low technology. Conversely, as mid-and high-tech solutions become more common, their intrinsic complexity is increasingly perceived as a major inconvenience, inefficient, overpriced, or inconvenient trait. Mid-tech assistive devices are electronic gadgets that require batteries to work and may or may not require training to use. Mid-tech Assistive devices used for learning include audiobooks, MP3 players, and other audio playback devices. Audiobooks and sound files may be useful if a person has difficulty reading printed materials. Individuals can utilize digital recorders as a mid-tech tool to take notes and enable the collection of audio to guarantee that no data is missing because of writing complications. Wheel-chair, Screen magnifiers, Gait trainers, Laser pointers, Voice amplifiers, Beep boxes, scooters, Braille translation software, Talking pedometer/watch, Switch adapted toys, Adapted seating, keyboards, Calculator, Electronic speller, etc., are the most well-known Mid Technology devices. Augmentative and Alternative Communication (AAC) devices rely on Mid technology, generally called speech-generating Devices (SGDs), which are typical battery-operated electronic devices and have simpler functions, whereas “high tech” AAC devices are electronic devices with highly advanced processors. High-tech AT devices, such as computers and specialised software, are more complex electrical devices that often include both hardware and software and include numerous functionalities to satisfy a variety of needs. High-tech AT is frequently used in conjunction with low-tech systems that can be employed in specific scenarios or as a backup in the event of a malfunction. Individuals with major functional impairments are the most common users of high technology, which is connected with almost necessary features but steep learning curves, complex restrictions and unpredictable results make it difficult to acquire, use, and maintain. High-tech assistive technology aids for writing include software, apps, and hardware devices. Text-to-speech, highlighting and notes, digital reading aids, idea mapping, word prediction, and a variety of other capabilities are all available in various literacy software packages. Evernote and Notability are note-taking apps that let users create an electronic notebook that can be shared and linked to reminders. The speech recognition software can transcribe an audio file into a text document when used in conjunction with a digital recorder. Smart pens, like the Livescribe pens, employ software combined with hardware to capture audio and

Assistive Devices

Advances in Data Science-Driven Technologies 11

sync it with notes on specific paper and note-taking apps like Evernote. Design Considerations for AT The design of the assistive device system looked to be tough even after several years of feedback. Although most assistive devices for the disabled appear to be easy, the technology that underpins their design and execution is usually fraught with complications. One of these challenges is that designers cannot depend on their own user experience since their capabilities vary markedly from those of disabled people. Development and Engineering for disabled individuals is a highly specialised object of research that examines how to establish a design strategy while working with impaired persons. There is no consensus in the scientific literature about which design technique is the most efficient when designing for impaired people or what aspects should be considered when picking the best acceptable design method. For designing assistive devices, the most extensively utilised techniques and tools are User-Centred Design (UCD) techniques and tools (modular elements of one distinct design method) [4, 5]. User-centered design (UCD) is a design strategy that stresses product adaptability to the user by involving the user throughout the design process [5, 6]. Many alternative approaches to AT design, such as iterative process, participatory design, emotion-driven design, USERfit & AD-SWOT & AD-TOWS, follow a UCD approach in some ways (such as involving users in various design stages), though the other design methodology is claimed. Evaluation of Functional Capabilities of Assistive Devices Assistive technology (AT) has a range or continuum of “low tech” to “high tech”. Both, the amount of technology as well as training required for the person who will be using it, decides whether the AT is low-tech or high-tech. Each piece of AT is tailored to the individual's needs and ability. The following steps are commonly involved in providing an AT evaluation after you and the individual determine the evaluation is necessary. Referral: As a service provider, we might request that an AT professional become involved in providing services. This typically involves the completion of a referral form, an in-person or telephone conversation about the person's needs

12 Advances in Data Science-Driven Technologies

Hiremath and C.R.

and the determination of the funding source. Scheduling and Evaluation: An evaluation often takes place on the job or in an educational setting. Evaluations can range in time from an hour or less in a single visit up to several visits over several hours. Report: The evaluator(s) generate a written list of their recommendations that typically includes specific equipment and services. In many cases, the individual can borrow trial equipment during and/or following the evaluation before the evaluator makes a final recommendation. Implementation: Once VR approves the recommendations and authorizes the funding for the equipment and any related costs, you will collaborate with the individual receiving services and your team to schedule and provide training and other services. You may need to arrange for ongoing technical support for a prescribed time. Possible Outcomes with AT People's attitudes regarding today's technology range from those who use it frequently and frequently and comfortably to those who use it rarely and with unease. Despite the fact that there are presumed positive correlations between AT utilization and living standard, non-use may not rule out obtaining a significant level of life expectancy. Those who believe capability enhancement in the context of a comprehensive assessment of a person's interests, needs, and life experience presume it is truly essential and cost-effective to improve a person's life quality, not just to restore functionality; to confront a person's emancipation necessities. Furthermore, the dynamic interplay between functional capacities, technological utilisation, and quality of life evolves with time. As a result, this complex of issues must be handled on a regular basis. The outcomes of an Assistive Technology Device framework begin when the device is purchased or placed in the consumer's hands [7, 8]. It progresses from short-term to longer-term outcomes such as Device Satisfaction, Effectiveness, Psychological Functioning, Efficiency, and Subjective, along with the influence of moderating factors such as personal and contextual factors. Environmental and Personal Factors are used by the AT components of Participation and Activities. All of these constructions are dynamic and recursive, which is crucial for the examination of ATDs. That is, they can modify and impact one another over time as a result of ATD use. As a result, they must be re-evaluated on a regular basis when looking at both short- and long-term outcomes.

Assistive Devices

Advances in Data Science-Driven Technologies 13

Feature Matching When it comes to matching a person with assistive equipment, there are numerous aspects to consider (Fig. 3). Irrespective of a category, a person is either a user or a non-user of the product under review. People, on the other hand, may differ within the classifications of use or non-use: use could be full-time and voluntary, or partial and hesitant; non-use can be due to completely avoiding or abandoning a gadget.

Fig. (3). Feature Matching for the Assistance.

An individual can simultaneously utilize any number of appliances that may belong to any number of categories. The introduction of any new device may replace the existing one, but this can create a situation where a user can find it more difficult to operate and adjust to the new features. Since it is known that a system made of many elementary supports might have an additive effect, resulting in a condition of overwhelmingly advanced for an individual, device configurability is expected to become a significant topic of concern as time goes on. The formal and informal process of determining the attributes of Assistive Technology are needed and desired as part of the AT Assessment Process. Ontology of Assistive Devices The concept of ontologies is critical for facilitating knowledge sharing and reuse. An ontology [9 - 11] is a formal description of concepts and relationships that really can emerge for a community of human and/or computer entities. An ontology is a characterization of a set of ideas in a shared domain [5] to facilitate the dissemination of information about technological products and best knowledge sharing, as well as set the way for a completely standardized evaluation procedure.

14 Advances in Data Science-Driven Technologies

Hiremath and C.R.

There have been no specific ontological matching initiatives for individualised preference portability across different programmes, platforms, and devices in the realm of assistive technology. However, the most important factor to consider when explaining the ontology of assistive technology is 'accessibility,' which ensures that people of various capabilities can interact with information and communication technologies (ICTs) [12]. Generally, the Ontology of the assistive devices can be formed on the basis of either the Matching Person and Technology (MPT) [4, 6] or Human-activity-assistive technology (HAAT [13 - 15]) models. The Human Activity Assistive Technology (HAAT) paradigm, which focuses on the relationship between the disabled observer and the facilitated activity in specific situations, is the most extensively used approach in the field of assistive technology. The Matching Person and Technology model [16] is quantified through a set of well-grounded and reasonable measurements which allow for a customized person-centered strategy to match people with the best solutions for particular needs. The Human Activity Assistive Technology (HAAT) paradigm represents any individual engaged in a certain activity while using assistive technology in a specific situation. The model's emphasis may be on an individual performing a task in a specific environment. As a consequence, any application of the framework begins with someone doing something in context, followed by the AT. The exercise, the individual, the Assistive Technology, and the context are the 4 elements of the HAAT model. The model's transactional aspect is supposed to portray an individual's experience when participating in activities, which contributes to AT by stressing the potential influence of the person's experience in a situation as he interacts with others and nonhuman elements. Situated knowledge is a notion that describes how a user's perceptions and interpretations of his current experience influence the current circumstance. The activity component of the HAAT paradigm assists the AT user in comprehending the activities for self-care, productivity, and recreation, which can be formulated in terms of time, space, and location. It directs product development, AT selection, and functional outcomes for evaluating AT use. The most critical part of the activity component is to determine the impact of AT use in diverse contexts. The human component is made up of the user's intellectual, physical, perceptual, and expressive capacities and requires a thorough understanding of human working nature in fundamental domains as well as the involvement of associated technological experience from a lifespan perspective. The term “context” refers to a physical environment, and the HAAT paradigm aligns with the social model of disability by making the contextual characteristics of Assistive Technology design, and service provision. Assistive technology is the final component of the HAAT model which comprises the human/technology

Assistive Devices

Advances in Data Science-Driven Technologies 15

interface (HTI), the processor, the environmental interface, and the activity output. The human/technology interface (HTI) acts as a conduit between humans and AT. The technology aids activity performance by enhancing cognitive manipulation, which is an activity output. The processor connects the HTI and the activity output to convert the knowledge into signals that regulate the activity output. This is accomplished by the environmental interface. The processor provides the interpreted data to the user through the HTI. Recognizing the activity assures that the gadget will assist in the completion of a productive task. The Matching person and technology form a person-centered System that usually identifies end-users preferences and needs to provide a service to match an individual's requirements. Most of the functions and features of the service model under this category are designed based on the past experience of users and their feedback which may undergo a series of paper-and-pencil measures. Hence, the Matching Person and Technology (MPT) process usually involves both personal and collaborative assessment [2, 5, 11]. This stage of the section- corresponds- to- Ontology-based Assistive- Devices (Fig. 4) and Applications that are among the finest in the industry [16]. Assistive devices have indeed been explored within commonplace, everyday contexts for more meaningful application using the Matching Person and Technology (MPT) and/or Human Activity Assistive Technology (HAAT) frameworks. Any individual's expectation can be reflected by their profile which may be implemented through ontologies. By the identification of such various common factors among the different contexts, in the domain of ontology of accessibility, the following aspects are explored.

Fig. (4). Ontology of Assistive Technology.

16 Advances in Data Science-Driven Technologies

Hiremath and C.R.

General Purpose Assistive Technologies Sensory Aids for Persons with Visual Impairment Assistive technology solutions for blindness and low vision include GPS devices with descriptive audio navigation, Smart magnifiers, Braille and Talking watches, Text-to-speech readers and spatial-aware mobility with IoT capability. The most prominent servicing solutions available include voice-command computer software for internet browsing and document management as well as bilingual display interpreting software with sophisticated resolutions and modified versions. Embossers, Braille displays, Keyboards and Printers with the desktop are the other services available. Sensory Aids for Persons with Auditory Impairment The application integrates development tools for assistive hearing aids, such as room-limited infrared systems, direct audio inputs, frequency modulation and inductive looping for deaf and hard-of-hearing people. Adjustments based on environmental conditions, Noise reduction algorithms, Remote controls, Automated settings and Bluetooth compatibility to access media audio and phone calls are all included in assistive listening systems and smart hearing aids. Enabling Function and Participation with Seating Technologies The degree of lineation, contouring and adjustability associated with each seating system facilitates the users by providing a balance between stability and mobility for the maintenance of neutral skeletal alignment to prevent skeletal deformities. Seating devices under this category have developed as discrete areas of intervention, each addressing the preferences of the target need to operate in all aspects of their life, such as resolving the issues related to postural control, tissue integrity and supporting a position to reduce user fatigue, enhancing the respiratory and circulatory function. Performance Areas Assistive Technology for Cognitive Augmentation Majority of the “cognitively” accessible technology is mostly a decade old. Cognition refers to the mental process of knowing, which includes features like consciousness, perception, thinking, and judgement, memory loss, dementia, language difficulties, the ability to make decisions, and the ability to operate independently are all symptoms of cognitive impairments. Specially designed technology for appropriate cognitive support (Fig. 5) can compensate for those cognitive impairments which can not be addressed by complex mainstream technologies [17]. The availability of soft technologies, such as appropriate selection, training, and implementation of technology solutions, is a critical element that makes the usage of Assistive technology for Cognition successful.

Assistive Devices

Advances in Data Science-Driven Technologies 17

Rehabilitation specialists, living skills counsellors and home health care providers [18] can all provide support to individuals with cognitive problems. Knowledge representation enables the proper interlinking of things, ideas, and events. The ability to recognise an object and the ability to recall the steps to do a task helps in figuring out the need for an assistive device to aid cognitive function. A Comprehensive evaluation of an individual's abilities relevant to the activity offers an exact estimate of the required support. Assistive technology selections are then moderated by the circumstance. Most of the commonplace technologies used for micro-prompting, alerting, storing and displaying, with reduced complexity and smart interfaces, produce output in the form of speech or text. Such devices also enable cognitive functions that render classifying products' functional categories more challenging.

Fig. (5). Assistive Technologies to Aid Cognitive Function.

Augmentative and Alternative Communication Systems The area of Augmentative and alternative communication deals with the complex communication needs of an individual [19]. Individuals who have difficulties in developing speech and language abilities can use augmentative and alternative communication devices to help them write and communicate. Many communicators rely on “No-technology” approaches such as speech, gestures, facial expressions, and vocalizations to communicate.

18 Advances in Data Science-Driven Technologies

Hiremath and C.R.

Low-tech systems in this category, such as paper communication boards and books, help meet communication needs quickly and easily, whereas high-tech devices, such as picture frames, smartphones, and computers, provide greater sophistication in available vocabulary, communication speed, and access flexibility. Speech-generating devices generate digitally recorded speech to help people with disabilities communicate more effectively. Assistive devices under this category help an individual to express their needs and transfer information. Technologies that Enable Mobility A person's mobility allows them to move to a location where the activity can be accomplished. Mobility with a wheelchair [20] is an activity that has garnered a lot of attention in terms of describing the abilities needed to be competent. The user must have a basic understanding of how to use brakes and manoeuvre in a wheelchair [21]. The degree of limitation in mobility determines the ambulation needs (Fig. 6). Individuals who are deemed marginal ambulators are at one extreme of the spectrum. Individuals with significant mobility restrictions who are reliant on manual mobility and for whom powered mobility is their only alternative for independence are on the other end of the spectrum.

Fig. (6). Scope of Mobility Limitations.

Assistive Technology for Manipulation and Control of the Environment

Manipulation is a term that describes the activities that we do with our upper extremities, specifically our fingers and hands. Using assistive devices, particularly ones that are electronic, demands a range of tactics. A person's everyday activities are the pinnacle of these manipulation components' integration. Special-purpose technology is required to suit the needs of Self-care, Recreation and Work. Assistive robotic systems are designed to help people with disabilities move items and operate independently by acting as a natural alternative manipulation device [22 - 24]. Robots are frequently used as personal assistants, with the purpose of assisting people with mobility impairments and/or intellectual limitations with manipulation. Everyday actions such as eating and personal hygiene are common tasks that are aided. Personal assistants are made up of stand-alone robotic arms to form robotic workstations and autonomous mobile robotic platforms that

Assistive Devices

Advances in Data Science-Driven Technologies 19

are integrated into a wheelchair.

Person-Centred care service with Assistive Technology New technologies for home comfort and assisted-living context support Relationship-centered care, Relational autonomy, Successful aging and Wellness to keep the resident safe [25]. Existing monitoring technologies such as wearables for health monitoring, fall alerts and other connecting devices make older adults' lives more comfortable [26, 27]. Technologies such as Robotics, Social management system, Information communication technology, Telecare and Virtual environments alleviate loneliness and social isolation. Residential and Nursing care facilities may also address the needs of those with dementia. Person-centred care service design promotes the use of technology in healthcare. Social robot designs can be used to serve the people with dementia. Though, Person-centered robotic design and automated systems under this category can mimic human sensing, cognitive behaviour and produce immediate human-like emotional responses; they are designed specifically to focus on individual users' perceptions, expectations and various behavioural and physical aspects. Personhood in dementia care has been integrated with socially supportive robots and context-sensitive computer approaches to create symbiotic robotic systems. Personal Emergency Response Systems Assistive devices for the Personal Emergency Response System [PERS] help the elders and the associated caregivers by offering reliable, quick access to emergency assistance [28]. Because of loss of consciousness or severe ailments, modern PERS is usually enabled by contacting emergency assistance to aid an individual in automatically activating the system. PERS with fall detection employ impact detectors and analytics to recognise whenever an individual has fallen and promptly call for help. In-home Monitoring with Assistive Technology Without the need for human interaction, in-home monitoring systems [29, 30] can examine users' everyday life routines and use that information to determine when assistance is required. Any unfavorable occurrences recognised in a home setting equipped with sensors are transmitted to a member of the family or a health care provider via activity analysis. These products are highly suitable to single-occupant families. Assistive Technologies in the Context of the Classroom On a variety of devices, computer-assisted schooling delivers instruction and practice possibilities. Using software systems, high-tech assistive solutions can be created for use in special education classrooms as well as for individual children with learning disabilities. Text-to-speech, Speech-to-text and word prediction programmes for desktops and Graphic Organizers for mobile devices, Pentop computers, Optical Character Recognition (OCR) software, Visuwords and Visual

20 Advances in Data Science-Driven Technologies

Hiremath and C.R.

Scene Display (VSD) platforms are some of the acceptable writing and reading options created with Universal Design for Learning (UDL) criteria. Assistive Technology in the Context of Workplace Assistive technology (AT) can help people with impairments overcome barriers to employment and work more productively. Over time, employees may acquire or develop disabilities. Companies that wish to help their employees retain their skills, talent, and experience as they age should invest in cutting-edge assistive technology. Appropriately configured technology can help AT users in the workplace by reducing the pain from awkward postures and movements. AT can also assist people with disabilities in adapting to new work environments and demands. Employing assistive technology gives businesses a competitive advantage by allowing them to hire from a bigger pool of eligible individuals without eliminating any. AT-enabled Office Settings Specifically designed Adapted, and Ergonomic keyboards can minimize strain and discomfort at the workplace. The content on the computer screen can be read by screen readers to produce it in a computerized voice, and Screen magnifiers make screen text larger for the person with low vision. Voice recognition software can help those with limited hand-use. Screen clips and Microphone headsets decrease neck strain. Issues Associated with Assistive Technology Practice The enhancement in individual independence is the main motivation for all assistive technological advances. The clinical decision-making process for a specific individual's assistive technology should therefore adhere to the professional and ethical principle of autonomy. The ethical principle of justice is concerned with the issue of fairness in individual, organisational, interpersonal and social situations, and it emphasizes individual liberty and choice. Despite its limitations, the assistive technology business meets the needs of highincome environments. In low-income countries, small-scale domestic manufacturers and distributors of assistive technology are often unable to address the requirements of everyone who needs it [31]. Furthermore, assistive technology facilities are sometimes scarce. Individuals with varied disabilities, ages, ethnicities, genders, and languages often do not have equal access to assistive infrastructure and supplies because they live in different regions of a country, and they live in a diverse state of the economy. Rising costs, restricted accessibility, a lack of knowledge, a scarcity of adequately trained workers, a paucity of administration, and insufficient funding for assistive technology all contribute to a lack of access [32].

Assistive Devices

Advances in Data Science-Driven Technologies 21

Assistive technology is provided by the government, religious agencies, overseas humanitarian relief, corporate, philanthropic and the private commercial sector. Due to the challenges in delivering a consistent supply of assistive devices and replacement parts due to a lack of regulations, logistical, finances, innovation, and expensive customs and excise costs, very few people receive access to a limited selection of assistive equipment. In low-income situations, assistive gadgets might be prohibitively expensive. Additional problems can include indirect charges and maintenance costs. Improper assistive devices can cause health issues thereby device disengagement. The effectiveness of any assistive device initiative hinges on the availability of adequate assistive device services. Assistive devices must be tailored to an individual's context and personal qualities to guarantee that there is a demand for them, that they are used, and that they are safe. For a variety of assistive devices, design standards have been developed. As practitioners and service suppliers are unaware of the variety of obtainable assistive devices and associated advantages, demands and incentives to supply are restrained. It is widely acknowledged that there is a great potential for assistive technologies, but provision is considerably lower, particularly in low- and middle-income nations. This imbalance between requirement and availability is a barrier to better access, and it is caused by a variety of circumstances, including a general lack of understanding between intended consumers, carers, and medical practitioners. In addition, there is a substantial unsatisfied requirement for assistive technologies in the treatment and care of illness and injuries - likewise, this high degree of requirement somehow doesn't correlate to a short supply. Overall, there is a dearth of information on the scale of the unfulfilled demand in this sector.The majority of research evaluating the usefulness of various forms of assistive devices leads to high environments, which is unexpected. A dearth of elevated, well-designed investigation in this domain has been emphasised by many studies synthesising outcomes to date. Attempts to Maximize the Accessibility and Affordability of Assistive Technology The potential for developing and manufacturing appropriate assistive gadgets at a reasonable price is enormous. The Worldwide Cooperation on Assistive Technology (GATE) and the WHO's Priority Assistive Devices List are working to promote access to affordable, high-quality assistive products on a global scale [32]. In addition, the WHO is assisting governments in the development of national assistive technology programmes. Adopting an integrated approach to assistive technology could make more economical options [33]. Market shaping could be used to lower transaction costs and balance suppliers, thereby consumer liabilities. The cheap cost and availability of glasses have aided a number of

22 Advances in Data Science-Driven Technologies

Hiremath and C.R.

activities in the eye health sector to promote access to affordable eyeglasses, such as through optical shops and school health programmes [34]. Community-based methods could help underserved groups gain access to assistive technology. Nonprofit organisations work to promote access to assistive devices by donating them to people who otherwise would not be able to buy them, albeit this model, which is based on donations, has sustainability difficulties. International governmental organisations, governments, non-governmental organisations, and the corporate sector may form partnerships to provide assistive equipment. Research Trends and Future Research Directions With the progress in smart technology and microsensors, wearable and other smart technologies have made it possible to remotely monitor the health of humans by closely tracking human activities without disturbing the user's motions. With the advancement of the Internet of Things (IoT) and smartphone technology, it is now possible to use assistive technologies to remotely monitor a patient's recovery. Wearable technologies can also be used to monitor hand joints and to help athletes recover from running-related ailments. Smart technologies encompass advanced materials utilised in protective gear and clothes, as well as electronic technologies. Wearable technology in rehabilitation gives high-quality care to a large number of people with complex medical conditions. Disabled people can participate in more activities with the help of remote monitoring devices, which also improves healthcare diagnostics. Many well-known multinational corporations are developing portable smart assistants for daily use that can accurately detect common health problems and provide real-time monitoring, eliminating the need for physical testing. Smart healthcare makes healthcare solutions more convenient and efficient by incorporating cutting-edge technologies such as big data, deep learning, artificial intelligence, IoT, and edge computing. Smart assistive technology improvements are targeted at assisting users not only in converting biological, personal, and environmental data into meaningful user cognition, but also in translating these insights into comprehensive judgments and goal-oriented activities. Microsensors, wearable gadgets, and other intelligent technologies have made it possible to remotely monitor people's health by carefully monitoring human activities without interfering with their movement. CONCLUSION The capability to live independently can deteriorate as people age. Various physical impairments as a consequence of aging in place, as well as physical impairments, necessitate assistance with personal care, aiding, restoration of trust, and self-esteem. The present study explores assistive technology facets,

Assistive Devices

Advances in Data Science-Driven Technologies 23

categories, and new-age technology support for the assistive devices adoption for the benefit of physically and cognitively challenged persons. It informs practitioners on the implications of how to create consistent care in the contemporary world in order to ensure long usage of assistive devices followed by subsequent and brief deployment, as well as how assistive devices can become a part of people's everyday life. The present analysis can be viewed as a first avenue in the realm of assistive technology as it emphasizes state-of-the-art assistive technologies. It also looks at assistive technology with characteristics tailored to the needs of the elderly. The study has included various aspects related to assistive technology for proposing public policies for the elderly. Using the ageing of the population and rising life expectancy as a preliminary step, the study emphasised smart self-management as a way of empowerment through the use of assistive technology, categories, and ontology. Performance areas of assistive technology are explored with a distinction made between assistive devices within various contexts. Many issues and challenges encountered by Assistive technology are also included. CONSENT FOR PUBLICATION Not applicable. CONFLICT OF INTEREST The author declares no conflict of interest, financial or otherwise. ACKNOWLEDGMENT We, the authors, would like to express our thanks to the esteemed editors for giving us the opportunity to explore this area of our research interests, which helped us in doing a lot of Research throughout which we attempted to reconnoitre many interesting paradigms related to the assistive technology domain. We gratefully acknowledge the support from the institution for providing the research environment with equipped facilities. We also thank the support from our respective families, who helped us to carry out this research work patiently. REFERENCES [1]

W. Elmannai, and K. Elleithy, Sensor-Based Assistive Devices for Visually-Impaired People: Current Status. Challenges, and Future Directions, 2017.

[2]

M. Jenko, Z. Matjačić, G. Vidmar, J. Bešter, M. Pogačnik, and A. Zupan, "A method for selection of appropriate assistive technology for computer access", Int. J. Rehabil. Res., vol. 33, no. 4, pp. 298-305, 2010. [http://dx.doi.org/10.1097/MRR.0b013e3283375e35] [PMID: 20216223]

[3]

S. Kirboyun, "High-Tech or Low-Tech? Impact of Assistive Technology in School Settings for Students with Visual Impairments: Review of Research", International Journal for Infonomics, vol.

24 Advances in Data Science-Driven Technologies

Hiremath and C.R.

13, no. 1, pp. 1945-1953, 2020. [http://dx.doi.org/10.20533/iji.1742.4712.2020.0201] [4]

C Magnier, Methods for Designing Assistive Devices Extracted from 16 Case Studies in Literature, 2012. [http://dx.doi.org/10.1007/s12008-012-0143-2]

[5]

C. Magnier, G. Thomann, F. Villeneuve, and P. Zwolinski, Investigation of methods for the design of assistive device: UCD and medical tools Proceedings of IDMME, 2010.

[6]

R.D Orpwood, Design methodology for aids for the disabled., vol. 14, no. 1, pp. 2-10, 1990. [http://dx.doi.org/10.3109/03091909009028756]

[7]

Desleigh de Jonge, and Wendy Stevens, Capturing the True Value of Assistive Technologies to Consumers in Routine Outcome Measurement., 2016. [http://dx.doi.org/10.3390/technologies4040035]

[8]

M.J. Scherer, "Outcomes of assistive technology use on quality of life", Disabil. Rehabil., vol. 18, no. 9, pp. 439-448, 1996. [http://dx.doi.org/10.3109/09638289609165907] [PMID: 8877302]

[9]

A. Danial-Saad, T. Kuflik, P.L. Tamar Weiss, and N. Schreuer, "Building an ontology for assistive technology using the Delphi method", Disabil. Rehabil. Assist. Technol., vol. 8, no. 4, pp. 275-286, 2013. [http://dx.doi.org/10.3109/17483107.2012.723238] [PMID: 23025744]

[10]

T.R. Gruber, "A translation approach to portable ontology specifications", Knowl. Acquis., vol. 5, no. 2, pp. 199-220, 1993. [http://dx.doi.org/10.1006/knac.1993.1008]

[11]

C Biihler, Approach to the analysis of user requirements in assistive technology., vol. 2, no. 17, pp. 187-192, 1996. [http://dx.doi.org/10.1016/0169-8141(95)00049-6]

[12]

Romero Marino, and Brunil Dalila, Accessibility and Activity-Centered Design for ICT Users: ACCESIBILITIC Ontology., 2018.

[13]

A.M. Cook, and J.M. Polgar, Activity, Human and Context. Assistive Technologies, 2015. [http://dx.doi.org/10.1016/B978-0-323-09631-7.00003-X]

[14]

International Classification of Functioning, Disability and Health – ICF. Word Health Organization, 2001.

[15]

International Classification of Functioning, Disability and Health – ICF. Word Health Organization, 2001.

[16]

Konstantinos Tsiakas, Maher Abujelala, and Fillia Makedon, Task Engagement as Personalization Feedback for Socially-Assistive Robots and Cognitive Training., 2018. [http://dx.doi.org/10.3390/technologies6020049]

[17]

Josef Wolfartsberger, Jean Haslwanter, and René Lindorfer, Perspectives on Assistive Systems for Manual Assembly Tasks in Industry, . [http://dx.doi.org/10.3390/technologies7010012]

[18]

B-J. Krings, and N. Weinberger, Assistant without a Master? Some Conceptual Implications of Assistive Robotics in HealthCare, 2018.

[19]

Y. Elsahar, S. Hu, K. Bouazza-Marouf, D. Kerr, and A. Mansor, "Augmentative and Alternative Communication (AAC) Advances: A Review of Configurations for Individuals with a Speech Disability", Sensors (Basel), vol. 19, no. 8, p. 1911, 2019. [http://dx.doi.org/10.3390/s19081911] [PMID: 31013673]

[20]

Domingues, Psychosocial Impact of Powered Wheelchair, Users Satisfaction and Their Relation to Social Participation., 2019.

Assistive Devices

Advances in Data Science-Driven Technologies 25

[21]

L. Maule, RoboEye, an Efficient. Reliable and Safe Semi-Autonomous Gaze Driven Wheelchair for Domestic Use, 2021.

[22]

Ahmad Lotfi, Caroline Langensiepen, and Salisu Yahaya, Socially Assistive Robotics: Robot Exercise Trainer for Older Adults, 2018. [http://dx.doi.org/10.3390/technologies6010032]

[23]

Ramviyas Parasuraman, and Byung-Cheol Min, Assistive Robotics, 2018.

[24]

K. Tsiakas, A Taxonomy in Robot-Assisted Training: Current Trends. Needs and Challenges, 2018. [http://dx.doi.org/10.1145/3197768.3197787]

[25]

J. Cahill, S. McLoughlin, and S. Wetherall, The Design of New Technology Supporting Wellbeing. Independence and Social Participation, for Older Adults Domiciled in Residential Homes and/or Assisted Living Communities, 2018.

[26]

B. Carrier, "Validity and Reliability of Physiological Data in Applied Settings Measured by Wearable Technology: A Rapid", Syst. Rev., 2020.

[27]

Mario Munoz-Organero, Editorial for the Special Issue' Personal Health and Wellbeing Intelligent Systems Based on Wearable and Mobile Technologies, 2018. [http://dx.doi.org/10.3390/technologies6010029]

[28]

V. Hessels, G.S. Le Prell, and W.C. Mann, "Advances in Personal Emergency Response and Detection Systems", Assist. Technol., vol. 23, no. 3, pp. 152-161, 2011. [http://dx.doi.org/10.1080/10400435.2011.588989]

[29]

R. Khosla, M.T. Chu, S.M.S. Khaksar, K. Nguyen, and T. Nishida, "Engagement and experience of older people with socially assistive robots in home care", Assist. Technol., vol. 33, no. 2, pp. 57-71, 2021. [http://dx.doi.org/10.1080/10400435.2019.1588805] [PMID: 31063044]

[30]

Evropi Stefanidi, ParlAmI: A Multimodal Approach for Programming Intelligent Environments, 2019. [http://dx.doi.org/10.3390/technologies7010011]

[31]

Brigitte Rohwerder, Assistive technologies in developing countries, 2018.

[32]

V. Tangcharoensathien, W. Witthayapipopsakul, S. Viriyathorn, and W. Patcharanarumol, "Improving access to assistive technologies: challenges and solutions in low- and middle-income countries", WHO South-East Asia J. Public Health, vol. 7, no. 2, pp. 84-89, 2018. [http://dx.doi.org/10.4103/2224-3151.239419] [PMID: 30136666]

[33]

Nandini Modi, and Jaiteg Singh, "A survey of research trends in assistive technologies using information modelling techniques", DISABILITY AND REHABILITATION: ASSISTIVE TECHNOLOGY..

[34]

H. Wahidin, J. Waycot, and S. Baker, "The Challenges in Adopting Assistive Technologies in the Workplace for People with Visual Impairments", In: Computer-Human Interaction Conference (OzCHI '18).ACM Press, New York, NY, 2018.

26

Advances in Data Science-Driven Technologies, 2023, 26-47

CHAPTER 2

Accessibility of Software/Hardware Meenu Chandel1,* and Manu Sood1 1

Department of Computer Science, Himachal Pradesh University, Shimla, India Abstract: The phenomenal growth in Information and Communication Technology (ICT) is rapid and is responsible for changing disruptively the way various day-to-day tasks were being performed earlier. A plethora of user categories has benefitted immensely from this upward growth. It is also providing society with a multitude of entertainment options. The support of user-friendly software platforms for various ICT applications and tools is crucial in all these activities. Unfortunately, in the past, the designers of many software and hardware systems have not appropriately considered the Persons with Disability (PwD) as the active co-fellows of this journey and are being left behind in most of such applications of ICT. Thus, this significant part of the world population often seems to be neglected. Accessibility to every user with specific reference here to the ICT has always been a very important issue. What may be easily accessible to a set of persons may not be completely or partially accessible to another set of persons with disabilities. In this chapter, we discuss various types of disabilities along with the accessibility of hardware and software. Further, we highlight the concept of web accessibility and ICT accessibility for PwDs.

Keywords: Accessibility, Accessible technology, Accessible Website, Adaptive technology, Assistive technology, Causes of impairment, Cognitive impairment, Communication difficulties, Handicap, Hardware accessibility, Hearing impairment, ICT, Persons with Disability, Physical impairment, Reading disabilities, Sensory impairment, Software accessibility, Visual impairment, Voice dictation system, Web accessibility. BACKGROUND Information and Communication Technology has seen a mammoth growth from the era of Industry 3.0 to 4.0 and still continuing into Industry 5.0. The developments in hardware, software and communication technologies, including the disruptive ones, continue to touch every part of the daily life of even a common man. As per the Internet Live Stats website, which is part of the Corresponding author Meenu Chandel: Dept. of Computer Science, Himachal Pradesh University, Shimla, India; E-mail: [email protected] *

Manoj Kumar M V, Immanuel Azaad Moonesar R.D., Ananth Rao, Pradeep N, Annappa, Sandeep Kautish and Vijayakumar Varadarajan (Eds.) All rights reserved-© 2023 Bentham Science Publishers

Accessibility

Advances in Data Science-Driven Technologies 27

Real-Time Statistics Project (Worldometers and 7 Billion World), on 1st November 2021 at 3:00 pm IST, there was approximately a) five billion internet users, b) 2 billion websites, c) 175 billion emails sent, d) 5 billion google searches, e) 5 billion videos viewed on YouTube, f) 2 billion active Facebook users, g) 380 million active Twitter users, h) 400 million active Pinterest users, i) 421 thousand computers sold, j) 2 million smartphones sold, k) 243 thousand tablets sold, l) 7 billion GB of Internet traffic, m) 2.9 million MWh of electricity consumed. Not only this, in one second, there were a) 9717 Tweets, b) 1112 Instagram photos upload, c) 1980 Tumblr posts, d) 6180 Skype calls, e) 97 thousand Google searches, f) 92 thousand YouTube videos viewed, g) 3 million emails sent and h) 132 thousand GB of Internet traffic on the net [1]. This portrays the latest snapshot of the length and breadth of penetration of the technology for normal human beings. But on the other side, some segments of the population, such as the elderly or those with disabilities, may have trouble accessing the new technology and services that society provides to the masses in general. Most of the software programmes overlook this disadvantaged set of users, and may be unintentional while presuming that all computer users can accomplish the following activities with ease and without any discomfort [2]: 1. Reading and responding to the text and visuals displayed on the screen. 2. Using the standard keyboard while keying in text/instructions. 3. Using the standard mouse to perform various operations on text, images and other data. 4. Paying attention to various audio signals and responding to them appropriately. There are several sections of individuals who face difficulties in performing one or more of the aforementioned activities and hence seem unable to access many prominent computer applications [3]. To operate a computing device, a user who is print impaired (e.g., blind, dyslexic, cognitively disabled, or illiterate), physically disabled with restricted mobility, or hearing impaired requires the support of some specific assistive technology. There is a kind of social barrier that at least restricts, if not prevents, a Person with Disabilities (PwD) from participating in particular activities or interacting with the environment around them. Disability is defined as a reduction in one or more of the following faculties of a human being: physical, cognitive, mental, sensory, emotional, developmental, or any combination of these. This reduction in the

28 Advances in Data Science-Driven Technologies

Chandel and Sood

faculties may be by birth or may develop over a period of time [4]. The medical fraternity categorises the causes of impairment as follows [3]: a. Through inheritance (genetically transmitted) b. Because of a congenital problem, infection or other diseases in the mother during pregnancy, an immature or deadly developmental abnormality, or an accident during or shortly after delivery c. Through the acquisition, such as problems brought on by an unidentified disease or accident after birth, anytime later This paper is an attempt to explore the role of accessibility of the Information and Communication Technology tools and infrastructure and the difficulties being encountered by various categories of PwDs while interacting with them. INTRODUCTION According to the Global Report on Disability, disability affects almost 15% of the total global population, which is approximately one billion people [3]. In nature, a handicap or a disability of a human being may either be clearly evident or invisible. Physically impaired members of society are unable to fully coordinate their physical motor abilities, resulting in restricted body motions, lack of body coordination, and/or decreased strength in various muscular structures. Visual impairment is the inability to perceive items as clearly as a healthy individual can. Near or far, vision impairment affects probably at least 2.2 billion people globally, which includes around 1 billion with vision impairment that might have been barred or rectified [5]. Visually Impaired (VI) persons have to strive harder, even in their own environments, to navigate from one location to the other and/or to locate objects around them. According to the World Health Organization (WHO), 253 million people are suffering from disabling vision impairment, out of which 36 million people are completely blind and 217 million are suffering from mild to moderate vision impairment [6]. Hearing impairment refers to a person's inability to hear words or sounds clearly and/or precisely. In such persons, any portion of the ear of a human being might be affected by inappropriate development, injuries, or infection(s). Hearing is a critical condition of appropriate speech and language development. Communication difficulties also prevent deaf individuals from socialising and working in the same manner that normal people do. Due to their limited communication access and engagement, deaf people frequently suffer damage and dissatisfaction in their personal and professional life [7]. The rest of the paper has been organized as follows. After covering the brief background and necessary introduction to the topic concerned, different categories

Accessibility

Advances in Data Science-Driven Technologies 29

of PwDs have been discussed with reference to the issue of accessibility. Next, a brief note on the hardware and software dimensions of the accessibility has been provided, which is followed by a detailed presentation of the role of assistive technologies in supporting the accessibility for the PwDs. In the subsequent sections, a cursor to disabilities and web accessibility and a snapshot of disabilities and ICT accessibility has been highlighted. And lastly, before concluding the paper, a few recommendations, as well as suggestions, have been laid down for enhancing the accessibility of ICT to the PwDs. ACCESSIBILITY FOR DIFFERENT CATEGORIES OF PWDS Accessibility is a broad term for referring to the degree of ease with which human beings, in general, acquire, use and apply knowledge for gaining some benefit [8]. The main purpose of software accessibility is to guarantee that the application software is installable and usable by any person as and when required with equivalent user experience. This necessitates that all the users be able to interpret and understand regularly, with ease and without difficulty, what is being presented to them and compulsorily includes using various controls [9]. Accessibility refers to how a product, equipment, service, or environment can be used and accessed by people of all abilities, including those with disabilities. Access to technical products, resources, and services spanning hardware and software is the major focus of digital accessibility [10]. Around 650 million individuals, or nearly 10% of the world's population, suffer from some form of disability, however small it may be. Many human beings are born with physical, sensory and/or cognitive impairment that makes daily chores more challenging for them to perform. Others may suffer from disabilities later in life due to some accident, sickness, or the natural course of ageing. Many elderly people are diagnosed with chronic illnesses leading to devastating disabilities. Visually Impaired Individuals There are numerous devices able to assist blind people. Some of them take the form of one-of-a-kind gear, refreshable Braille displays, portable CCTVs, and large-print keyboards. Furthermore, the software seems to be the essential integral of any such technology. Blind people, for example, frequently use screen reader software such as JAWS [11]. For surfing and navigating a computer, several major operating systems, such as Windows 10 and OS X, may give a synthetic voice. Several apps are now available to enable blind users “see” and navigate various computing systems more effectively. The most recent smartphone applications can identify items from photographs audibly or even provide realtime assistance from sighted volunteers when necessary.

30 Advances in Data Science-Driven Technologies

Chandel and Sood

Physically Challenged Individuals There are a lot of innovations that focus on providing improved input devices when it comes to supporting easy mobility to physically challenged people. The majority of individuals are able to comfortably operate a keyboard and/or a mouse, but others with conditions like paralysis, missing limbs, or neuromuscular disorders may find it challenging. As a result, depending on the degree of difficulty in operating the input devices, support from alternative gadgets may be offered for such individuals [12]. A trackball or joystick may be more effective for persons with less serious physical difficulties. Options to operate a mouse with head movement, eye movements, and sip-and-puff gestures are also available for those with greater difficulty (e.g., Windows 10 supports eye-control devices). For people who only have one hand, very efficient one-handed keyboards also offer an opportunity. Deaf and/or Hearing Impaired Individuals Since most interaction with gadgets and other computing devices is based on the ability to read and write, deaf individuals can embrace computers and the Internet more effectively than other groups. A growing number of gadgets and applications with Assistive Technology (AT) for deaf individuals are being developed. Many of them are dedicated to helping deaf individuals with common problems. Apps that translate voice to text in real-time or create subtitles in many languages, as well as gadgets that turn auditory cues into physical feedback, are really quite popular. HARDWARE AND SOFTWARE ACCESSIBILITY FOR PWDS People with substantial physical limitations may have trouble using a regular mouse and keyboard to use their computers. They could also have trouble operating electronic things [13]. Assistive technologies are available, which is the most impactful. This is contingent on the individual's personal demands and the things they desire to be capable of completing independently. What kinds of technology may persons with physical limitations benefit from? There are a variety of hardware and software choices available, which may be used independently or in combination. Hardware Options Given below are some of the alternative hardware options available for different classes of the PwDs.

Accessibility ●

Advances in Data Science-Driven Technologies 31

Alternatives to the standard mouse Trackballs Joysticks Controlling the mouse with a switch and any trustworthy body part movement Only head movement is used to control the mouse Only eye movement is used to control the mouse Mouse control via sip/puff mouth movements Alternative keyboards are capable of being programmed, handled with one hand, or just providing larger, easier-to-select keys. Key guards are positioned on top of the keyboard to make it easier to choose the desired key without activating others. Automatic page-turners aid in the reading of physical books or magazines Additional technology, such as voice-producing devices, can help people with physical limitations with additional speech issues related to independently communicating. Integrated powered mobility controls enable power wheelchair controls to be utilised not only to manage the wheelchair itself, but also to operate computing and/or a speech-generating device. ❍ ❍ ❍

❍ ❍ ❍

●

●

● ●

●

Software Options Some of the accessibility options available to different categories of PwDs through software are: ●

●

●

●

● ●

●

Voice recognition software turns speech into text without needing to use a keyboard or mouse. Word prediction software dramatically reduces the number of keystrokes required to type text. Computer access software allows users with significant disabilities to operate any computer program with a switch or multiple switches. Switch training programs assist younger students in becoming proficient in switch use. One-handed typing tutor. Screen enlargement software helps the users who suffer from partial visual impairments up to some extent. Software for converting print documents to electronic versions helps in avoiding the need to be able to turn pages or move papers.

32 Advances in Data Science-Driven Technologies

Chandel and Sood

ASSISTIVE TECHNOLOGY Adaptive or assistive technologies are in the form of hardware and software tools that have been created to provide functional alternatives to some of the regular processes within the realms of ICT. Many persons with disabilities have a number of difficulties when it comes to giving input to a computing device, understanding the output, and/or reading the documentation part [14]. People with impairments heavily depend upon suitable keyboards, mouse, and display screens; thus customised hardware and software are being created for this purpose. Software for physically disabled persons has a distinct niche among the numerous software options in the market. The application field of such software is fairly narrow, yet this does not diminish the demand for delivering adaptable software for persons with physical impairments. All kinds of customers, including those with restricted physical capabilities, are allowed to use the freedom provided by access to the Global Web. According to Planet Health Organization (PHO), 253 million people have a vision problems, 36 million of them are blind, and 217 million have moderate to severe vision loss [15]. In a nutshell, computer program design should accommodate the capability to see, listen, perform inputs, read the text, or process information ranging from one user to the next through time, depending on the utilization circumstances [16]. As a corollary, ICT products and services need to be designed to be accessible to as many users as possible. Gadgets, tools, hardware, and software that enable people with disabilities to access computers are referred to as assistive or enabling technology. It provides an alternative approach for retrieving screen content, controlling the computer, and entering information. The following are examples of specific computer modification software or devices [17]: 1. Screen reading software (speaks displayed text and permits keyboard simulation and mouse operations) 2. Screen magnification software (enlarges screen contents) 3. Braille display (displays Braille letters), alternate input devices (e.g., screen keyboard) and special keyboards (to make data entry easier) 4. Mnemonics and shortcut keys for the keyboard (including Sticky Keys, Mouse Keys, Repeat Keys, Slow Keys, Bounce Keys or Toggle Keys) 5. Alternate pointing devices (such as foot-driven mice, head-mounted pointing devices, or eye-tracking systems)

Accessibility

Advances in Data Science-Driven Technologies 33

6. Software mouse simulators (for moving the mouse pointer by pressing keys on the numerical keypad) 7. Comprehension software (allows a dyslexic or learning-disabled computer user to see and hear the text as it is manipulated on the computer screen) 8. Predictive dictionary (speeds up typing by recommending words as the user types) Depending on the kind and degree of the disability, the above-mentioned technologies are intended to be employed in the situation of mobility, vision, hearing, and/or brain disorders. Low-vision computer users may benefit from screen magnification software because it enables them to understand sections of the screen more easily by magnifying the screen's information. It may be extremely important for physically disabled people to use a keyboard or mouse in addition to a standard keyboard or mouse to use the computer. Many people with disabilities use computers without the use of assistive devices, but they navigate 1025 using the keyboard [18]. Others enhance their interactions using hardware add-ons such as different pointing devices. The normal persons in their everyday life routines hardly realize how the impaired people live their life or the challenges they face in society. Their lives are mostly restricted to their homes and/or offices and/or social set ups, and their loops of interaction are thus confined to their individual social circles only. Under such circumstances, the sufferings being faced by the PwDs remain largely ignored. They are compelled to lead lives with minuscule support from society at most places across the globe. The role ICT hence assumes a great significance under such circumstances. Many of the roadblocks being encountered by PwDs thus can be removed, and the sufferings of such persons are minimized with the help of developing technologies at various fronts. Using computing technology, students and staff with disabilities can manage a wider range of activities independently, such as reading and writing papers, connecting with others, and searching the Internet for information. The digital era has ushered in a slew of technological advancements that essentially benefit the community of disabled persons [19]. Technology that enhances accessibility for individuals with impairments often falls into the following three broad areas, as outlined below: a. Assistive technology encompasses a set of specific-purpose devices designed to improve/enhance the functional abilities of disabled people.

34 Advances in Data Science-Driven Technologies

Chandel and Sood

b. Adaptive technology is a mechanism that permits people with disabilities to use technology that might rather be otherwise inaccessible to them. c. Accessible technology comes with a wide range of innovative and enhanced user experiences that help in breaking down some of the barriers being faced by PwDs and making their world a more welcoming place. But there are several limitations that can make successful use of such technologies difficult. These limitations, which can be attributed to the impairments acquired as a result of some disease(s) or to congenital impairments, include, but are not limited to: i. Congenital impairments such as Learning problems, dyslexia, and congenital impairments, head injuries, autism, developmental disabilities. ii. Visual impairments such as poor vision, blindness (total or partial), and colour blindness. iii. Hearing-related disabilities such as deafness, hearing loss, or hyperacusis. iv. Motor or dexterity impairments such as paralysis, spastic paralysis, dyspraxia, carpal tunnel syndrome, and repetitive strain injury. Persons with such impairments confront a variety of challenges when it comes to using technology, especially computing devices. These impediments are frequently classified into three functional groups [20]: 1. Difficulties in entering data into a computer 2. Interpreting the results 3. Examining supplementary materials It is disheartening to note that most of the PwDs continue to encounter challenges in accessing the full range of technology-based opportunities accessible to nondisabled people. Challenges to standard computer software, in particular, hinder educational and job possibilities for some persons with impairments. Deaf pupils, for instance, are still unable to understand a segment of a multimedia presentation that utilizes voice delivery without captions or translation. Similarly, an academic tutorial software that necessitates the use of a mouse is inaccessible to a student who is unable to utilise this instrument. To use software applications, some of the impaired people require specialised software and hardware that are covered under the category of assistive technology. A blind person, for example, might very well utilise a screen reader application with a voice synthesiser to access a program's information and capabilities. This system enhancement allows blind users to access text on the

Accessibility

Advances in Data Science-Driven Technologies 35

screen as well as keyboard operations, but it does not allow them to examine graphics or utilise functions that require the use of a mouse [21]. To guarantee all possible users have seamless access to a plethora of modern ICT tools, software developers must avoid the unintentional creation of barriers for persons with disabilities [22]. They must resolve to create solutions using assistive technology that is interoperable for both impaired as well as non-impaired people. In the history of social interactions, computer accessibility refers to the accessibility of a computing system to any individual, regardless of her/his degree of disability. An assistive system is a means that can aid those who are unable to talk audibly on their own. The system may employ speech, gesture, sign language, signals, synthesized speech, specialised communication devices or microcomputers. Some of the instances of assistive technology are [23]: i. Optical Character Recognition (OCR) software packages that are commonly used to scan printed items straight into the PC to meet a wide range of disabilities ii. Screen readers are software programmes that output either voice or Braille and are often used by blind or visually impaired persons iii. Non-assistive computer systems such as electronic messaging and instant messaging enable persons with hearing impairments to communicate over the Internet iv. Magnification software that enhances the size of text and visuals on computer monitors and is commonly used by those who have impaired vision or difficulty in reading To manage computers and dictate documents, voice recognition and dictation system are two powerful assistive technologies that enable people with disability to use spoken instructions. Most computer manufacturers assist people with impairments by including accessibility features in their operating systems, and examples are Microsoft Windows, Apple Macintosh, and UNIX. Educational software that delivers multisensory experiences, interaction, positive pillars, customised teaching, and repetition might be beneficial in skill development. Some children with learning impairments who have trouble processing textual material might benefit from using computers to complete writing projects, tutorial lessons, and drill-and-practice work. A normal word processor, for example, can be a useful tool for those with dysgraphia, or the inability to generate consistent handwriting. Some technologies make it possible for those who can use their hands, either partially or not at all, to utilise a conventional keyboard. Individuals with one finger or access to a mouth-stick, head-stick, or other pointing devices can operate

36 Advances in Data Science-Driven Technologies

Chandel and Sood

the computer by pushing keys with the pointing device. Software tools can generate “sticky keys” which electronically latch the SHIFT, CONTROL and other keys, allowing successive keystrokes to submit instructions that would ordinarily need two or more keys to be pushed simultaneously [24]. For individuals who are unable to release a key quickly enough to avoid numerous entries, the key repeat feature can be deactivated. Those with impaired fine motor control can wear keyboard guards (solid templates with holes over each key to aid in accurate selection). It is difficult to educate pupils who have a learning impairment. Because of their short attention span, they are quickly distracted. To guarantee an effective learning session, a learning system that can replicate their interest and capture their attention is required. Augmented Reality, a teaching aid assistive technology, combines a virtual world with a real one to make learning more enjoyable and interactive for children with learning impairments [25]. DISABILITIES AND WEB ACCESSIBILITY The web's power lies in its universality, remarked Tim Berners-Lee, Director of the World Wide Web Consortium (W3C) and creator of the World Wide Web. Access is critical for everyone, regardless of disability [26]. An increasing percentage of websites, tools, and technology are being developed in such a manner that people with disabilities can use them so as to a) observe, Perceive, Navigate, and Engage with the Web; and b) contribute to the Web. There are four types of limitations that impact a person's opportunity to utilize the Internet [20]: i. Difficulty to move, inadequate dexterity to use a mouse or keyboard, inability to regulate undesired movement, and limb deficiency ii. Hearing loss, which can range from partial hearing to complete deafness iii. Impaired vision, including colour blindness and partial or complete blindness iv. Issues with cognition and learning, also including reading, understanding, staying focused, remembering, and writing Some of the prominent impairments that otherwise prevent people from using the Internet but are being catered to nowadays include: a. b. c. d. e. f.

Auditory Cognitive Neurological Physical Speech Visual

Accessibility

Advances in Data Science-Driven Technologies 37

Web accessibility has been linked to all of these above-mentioned impairments. The W3C designed the Web Content Accessibility Guidelines (WCAG) 2.0 and referred to web design features that allow users to see, understand, operate, and participate in technology on websites. The WCAG is a set of guidelines for web designers and developers that aims at reducing accessibility problems [27]. The practise of making websites accessible to everybody, especially people with impairments, is known as web accessibility. As the Internet is becoming an increasingly important element of post-secondary education, it is vital that instructional websites be built to be accessible to disabled students. The process of creating websites accessible to persons who need features more than those provided by standard web browsers so as to utilize the Internet is the underlying principle for web accessibility. In a classroom context, an “accessible Web site” is one that enables students to access information in a variety of ways. Most of the web site built with the objective of artistic direction rather than universal distribution in mind. According to Carter and Markel, only one percent of web developers considered accessibility while creating web sites [28]. When web sites are really not designed with disabled people in mind, impediments to access the site's content are commonly absent. Images without alternate text, inaccurate use of structural components on a web page, unedited audio or unidentified video, generalized linear tables that are hard to understand, and websites with poor desaturated colours are all exchanging information obstacles. The World Wide Web Consortium is responsible for the vast majority of web accessibility standards (W3C). 3 The World Wide Web Consortium (W3C) is indeed an international Organisation were members, full-time staff, and the general public collaborate to establish web standards. The World Wide Web Consortium (W3C) is the leading authority on website standards, guidelines, software and tools. The Web Content Accessibility Proposal was established by the W3C in the 1990s (WAI). The Web Content Accessibility Guidelines (WCAG 1.0) were established by the WAI and were modified in December 2008 by WCAG 2.0. As WCAG 2.0 is still pretty recent, many authoring and assessment tools, as well as Section 508's legislative necessity, are still tailored around WCAG 1.0. In addition to WAI standards, the International Organization for Standardization (ISO) has also created web accessibility standards [29]. Web accessibility guidelines are also published by the ISO. ISO 9241 (titled Ergonomics of Human System Interaction) is the most applicable for this subject. It is a compilation of 28 sections (System Concepts, 2009). Initially, the internet helped many challenged participants to complete new duties. It still does; however, issues have occurred, primarily as a result of web developers' increased usage of sophisticated multimedia and design features in their sites. “As more sites offer interactive visuals and images, and home page clicking on a graphic is the only way to go from page to page, the online is not a favourable environment

38 Advances in Data Science-Driven Technologies

Chandel and Sood

for the visually challenged” [30]. On a typical homepage, there are various hurdles to seamless accessibility. Seven common accessibility barriers are listed by the World Wide Web Consortium [31]: ● ● ● ● ●

● ●

Photos that do not have appropriate text Hotspots in image maps with no alternatives text The use of structural components on a page that is manipulative Audio that is not captioned or video that is not described Lack of alternatives information for users who are unable to view frames or scripts Linear system tables that are difficult to read Sites with insufficient colour contrast

The need of the hour is that the Internet should increasingly strive to achieve higher degrees of inclusiveness as technology advances. If the web designers fail to integrate the concept of universal design while building the websites, a large proportion of participants would be unable to participate and contribute. According to Waddell, a California attorney and prominent promoter of online accessibility, “the growth and success of the emerging digital economies require that specific attention must be paid to the mechanism for permitting dynamic involvement of all sections of the society” [32]. Similar sentiments have been reflected through a study in Poland which has suggested that the degree of digital divide being faced by the PwDs today is really high and worrisome [33]. DISABILITIES AND ICT ACCESSIBILITY The rapid development, coupled with the widespread adoption of ICT, has fundamentally changed almost every aspect of life. For example, in education, ICT has changed the very way of accessing and utilising teaching, learning and research resources. Considering the growing availability of technological advances and advanced assistive devices, ICT is a potent tool to enable alternative options for learning by visually impaired students [34]. At the same time, if not supported with the concept of inclusiveness, the same ICT can become a nightmare for some sections of the persons with disabilities flipping accessibility with inaccessibility [35]. The following are some examples of inaccessibility: 1. Automated Teller Machine (ATM) at a bank may be inaccessible in the following ways. The machine is just too high for a wheelchair user to reach some of the controls. Low contrast between text and backdrop, or text-only visual

Accessibility

Advances in Data Science-Driven Technologies 39

display, making reading difficult or impossible for people with eyesight or reading disabilities 2. Responses and replies in an ATM are expressed in challenging terminology or complicated language, making them difficult for those with cognitive or learning impairments to understand [36]. 3. Websites with a mix of text, images, links, buttons, tables, interactive forms, and other content can be inaccessible in the following ways: a. On-screen icons are developed to adapt only to a click of the mouse, so a person with a physical disability who cannot use a mouse could perhaps ‘click' buttons by pressing the ‘Enter' key on their keyboard. b. The labels of input boxes and buttons on a payment form (e.g., name, choice of payment method) are presented in a way that a blind individual's text-to-speech software is unable to read, thus the blind person is unclear about the function of each box or control. 4. For a control, instructions or output to be fully accessible, all users must be able to do the following three things [37]: a. Perceive: Be aware of its existence and be able to obtain its contents. A deaf person, for example, may be ignorant of the presence of an auditory alarm signal, whereas a blind person may be unaware of the existence of a visual signal. b. Understand: Know what that means and how to use it. A person with a learning disability, for example, may be unable to understand the difficult or poorly worded ATM instructions. This can lead to hard times for that person in figuring out where to look for help. c. Operate: Be able to reach it and physically engage with it in order to complete the task, which may include pressing, moving, twisting, or tugging. A wheelchair user, for example, may be unable to reach an ATM's card slot. A blind individual will be unable to choose a location on a map. Fig. (1) depicts the accessibility statistics for PwDs in terms of the places where ICT facilities are generally used and the frequency of their usage [38]. If the ICT industrial sector does not include accessible design in their product development cycles and has no motivation to do so, accessible ICT and service demands for people with disabilities can never be addressed. It may not happen without addition financial consideration from the respective business houses [37].

40 Advances in Data Science-Driven Technologies

Chandel and Sood

ICT Access Location/Place

ICT Accessibility Data of PwDs

47.4

Computer Liberary

9 36.8

Internet Café

7

%age of PwD Users 15.8

Home

No. of PwD Users

3 0

10 20 30 40 50 PwD Users

Fig. (1). Places/Locations for accessing the ICT facilities.

Accessibility features that are the most desired by persons with disabilities as per the individual needs include: 1. Large print capability 2. Clear audio messages with precise interpretation to the person 3. Legible text messaging 4. Option for connecting the hearing aids 5. Suitable audio amplifications 6. Keys with large characters/Figs having good contrast ratio 7. High-quality videos, suitable for sign language, lip reading and person recognition 8. Interoperable IP and 3G networks 9. Real-time text generator using voice with sufficient tolerance to variations 10. Flashing light on incoming calls 11. Vibration on incoming calls 12. Video relay for sign languages 13. Captioned telephony relay services

Accessibility

Advances in Data Science-Driven Technologies 41

Frequency of Using ICT Facilities Mozilla Firefox, in many of its ongoing continuous evaluation projects, has posted some interesting data about the usage of web services by 19 Persons with Disabilities. The participants were asked how frequently visually impaired students used the accessible ICT facilities to improve the quality of their learning opportunities. As indicated in Fig. (2), eight (42.1%) of the participating visuallyimpaired students used ICTs very regularly, six (31.5%) used them frequently, and five (26.3%) used them just sometimes [40]. ICT Facilities Usage

Very Often, 8, 42%

Sometimes, 5, 26%

Often, 6, 32%

Fig. (2). Frequency of using ICT facilities [40].

Challenges Constraining Access to and Use of ICTs by the PwD Some of the key obstacles experienced by the PwDs while accessing and using ICT facilities in learning under UDSM's Special Education Unit were highlighted in research undertaken by some experts. Fig. (3) shows that 28 (77.7%) of them pointed towards ‘inadequate friendly ICT facilities’, 20 (55.5%) attributed the difficulties to ‘ineffective ICTs training provisions’, six (16.6%) identified ‘power cut offs’ as impediment, four (11.1%) regarded ‘outdated ICT facilities’ as a barrier, three (8.3%) mentioned a ‘shortage of ICT technicians’ for repairing ICT facilities as obstacles, and two (5.5%) cited poor Internet connectivity as the reason for difficulties [41]. Inadequate Friendliness Fig. (3) shows that the majority of respondents (77.7%) listed poor ICT facilities as the major obstacle the PwD students experience while accessing and utilising ICT infrastructure for learning. In this regard, one of the respondents stated: “The Special Education Unit's ICT facilities are insufficient to enable effective learning” [39]. Learning flexibility is limited as a result of this.

42 Advances in Data Science-Driven Technologies

Chandel and Sood

Impeding Factors

Factor Analysis of Opinions of PwDs Inadequate Friendly ICT Facilities Ineffective ICT Training… Power Cut-Offs Outdated ICT Facilities Shortage of ICT Technicians Poor Internet Connection 0

44.4

28 31.7

20 6 9.5 4 6.3 34.8 23.2 10

20

30

40

50

No. and %age of PwDs %age

Frequency

Fig. (3). Challenges constraining the use of ICTs.

Ineffective Training Provisions Considering the significance of training in improving skills and knowledge, the UDSM's Special Education Unit did not effectively provide training on the use of ICT tools/devices and reading abilities. According to this data, 55.5 percent of those polled believed that the Special Education Unit's training was unsatisfactory. Power Supply Outages For ICT facilities to function seamlessly, one of the most important ingredients required is a constant power source. Power outages and unpredictable energy supply were also mentioned as a challenge in this study that visually impaired students encountered when using ICT facilities. Due to their disruptive characteristics, power outages have been proven to hinder the proper implementation of facilities and degrade the learning process. During data collecting at the Special Education Unit, the researchers experienced one such power outage. The existing generator is not automated. Manually switched generators often require a man to manage them, which may not be adequate to safeguard against the disturbance of learning activities caused by power outages, especially when emergency power is not turned on promptly enough. 16.6% of the respondents agreed to choose this factor as a barrier to learning for PwDs. Outdated ICT Infrastructure The ICT infrastructure is continuously evolving through new technologies and innovations. Hence, there are ample chances of the ICT tools and devices getting

Accessibility

Advances in Data Science-Driven Technologies 43

outdated before the PwD users get accustomed to them. 11.1% of the respondents attributed the difficulties faced by them to this factor. Shortage of ICTs Experts and Technicians One of the challenges impaired students faced was the lack of specialists and experienced technicians for a) fixing the operational/maintenance issues faced by ICT facilities and b) innovating various applications to enhance the learning experience of impaired students. 8.3% of the respondents blamed this factor as a barrier. Internet Connectivity Mostly, the seamless Internet connection is considered essential for connecting different important formats for facilitating successful learning. The poor Internet connectivity at the UDSM negatively influenced the impaired students during their learning activities. Only 5.5% of them identified it as the main difficulty. Shortcomings in ICT facilities faced by PwDs is shown in Fig. (4). Shortcomings Faced by PwDs Lack of wide access to learning material, 5, 13% Depends on…

Time consuming, 28, 72%

Lack of wide access to learning material

Fig. (4). Shortcomings in ICT facilities faced by PwDs [42].

Results of Shortage of ICT Facilities In a study conducted by Bergman and Johnson, the authors have highlighted some of the significant difficulties faced by visually impaired persons. Visually challenged individual’s learning is hampered by a lack of access to user-friendly ICTs. According to the statistical results, 28 (77.7%) of the respondents mentioned time consumption, six (16.7%) reported over-dependence on the readers, and five (13.9%) cited lack of extensive access to learning resources as the main reason that hampers their learning processes [43].

44 Advances in Data Science-Driven Technologies

Chandel and Sood

RECOMMENDATIONS AND SUGGESTIONS Various categories of PwDs suffer in isolation or with their near and dear ones because of their different impairments. The role of ICT, as well as accessibility, in making their lives better is too huge to be ignored anymore. Various bodies involved in uplifting the living standards of such people have been putting their efforts tirelessly. One of the most significant recent initiatives has come from International Telecommunication Union, an agency of the UN with its headquarters in Geneva. Under its Study Group 1, Question 7 initiative, this body has made some meaningful recommendations and suggestions based upon its studies during the period 2018-2021 that have been listed here for ready reference. These are: a) overhauling of existing policies, regulations, legislations etc., related to ICT, b) in-depth interfacing with PwDs or their representative bodies for consultations and due approvals of such changes, c) Spreading awareness among PwDs about the existing provisions and the suggested changes after approval related to the accessibility to ICT, d) Large scale adoption of related prescribed norms/ standards/ guidelines regarding QoS and other technical standards, e) Revisiting the definitions and related legislations periodically, f) provisioning for impairment specific QoS and other ICT services for such people [44, 45]. CONCLUSION Although individuals with disabilities comprise a significant and diverse subset of the entire population on this earth, this subset is generally overlooked, maybe obliviously. Their needs and expectations are not getting sufficiently aligned with the normal requirements of the software and hardware designs and their subsequent advancements. Much of the problem arises due to the lack of awareness about the general-purpose as well as disability-specific needs of the PwDs, although designing accessible hardware/software makes them more usable for all such subsets of users. Accessibility is a critical issue for facilitating the assurance of equal opportunities not only for people with disabilities, but also for the plethora of other user categories. Hence, the accessibility inclusively needs to be aimed at improving the user interfaces and simplification of ICT operations for such people. It has been clear from what has been presented in this paper that there is an urgent need to address this issue so that persons with all kinds of specific disabilities can also use ICT to their advantage. The challenges and the specific solutions in reference to accessibility for each type of disability/impairment can be further taken up as future work. The study offers a number of recommendations for the parent institution to increase access to and use of ICT facilities to assist the PWD's learning activities, based on the study's results and conclusion.

Accessibility

Advances in Data Science-Driven Technologies 45

CONSENT FOR PUBLICATION Not applicable. CONFLICT OF INTEREST The author declares no conflict of interest, financial or otherwise. ACKNOWLEDGEMENT Declared none. REFRENCES [1]

Accessed: Nov. 1, 2021. Web site [Online]. Available, https://www.internetlivestats.com/

[2]

A. Newell, "Inclusive design or assistive technology", In: Inclusive DesignLondon, UK: Springer, 2003, pp. 172-181. [http://dx.doi.org/10.1007/978-1-4471-0001-0_11]

[3]

World Health Organization, World report on disability, 2011.https://www.who.int/teams/ noncommunicable-diseases/sensory-functions-disability-and-rehabilitation/world-report-on-disabilit

[4]

D. Blackwell, J. Lucas, and T. Clarke, "Summary health statistics for U.S. adults: National Health Interview Survey", Vital & Health Stat, National Centre for Health Statistics, ser. 10, no. 260, p. 2014, 2012.

[5]

M.A. Hersh, M.A. Johnson, Ed., Assistive Technology for Visually Impaired and Blind People. Springer-Verlag, Ltd: London, UK, 2008. [http://dx.doi.org/10.1007/978-1-84628-867-8]

[6]

M. Alnfiai, and S. Sampali, "Social and Communication Apps for the Deaf and Hearing Impaired", International Conference on Computer and Applications (ICCA), 2017pp. 120-126 [http://dx.doi.org/10.1109/COMAPP.2017.8079756]

[7]

P. Strumillo, "Electronic interfaces aiding the visually impaired in environmental access, mobility and navigation", 3rd International Conference on Human System Interaction, 2010pp. 17-24 [http://dx.doi.org/10.1109/HSI.2010.5514595]

[8]

M.A. Hersh, and M.A. Johnson, Accessible Environments.Assistive Technology for Visually Impaired and Blind People.London, U.K.: Springer-Verlag, 2010, pp. 323-361. [http://dx.doi.org/10.1007/978-1-84628-867-8_10]

[9]

M. Paciello, People with disabilities can’t access the web., 2000.http://xml.coverpages.org /pacielloDesigning.html

[10]

W. Chishom, Enabling your web site: A brief introduction to disabilities affecting web use., 2000.http://designshops.com/pace/ds/pub/1999/08/able.html

[11]

L. Valdes, Accessibility on the Internet., 1999.http://www.un.org/esa/socdev/enable/disacc00.htm

[12]

P. Zervas, V. Kardaras, and D.G. Sampson, "An online educational portal for supporting open access to teaching and learning of people with disabilities", 2014 IEEE 14th International Conference on Advanced Learning Technologies, pp. 564-565, 2014. [http://dx.doi.org/10.1109/ICALT.2014.165]

[13]

H. Ben Brahim, M.K. Khribi, and M. Jemni, "Towards accessible open educational resources: overview and challenges", 2017 6th International Conference on Information and Communication Technology and Accessibility (ICTA), pp. 1-6, 2017.

46 Advances in Data Science-Driven Technologies

Chandel and Sood

[http://dx.doi.org/10.1109/ICTA.2017.8336068] [14]

R. Reji, R. Damodar, R. Pillai, S. Nair, and M. Deshmukh, "Intelligent reading assistant for the visually impaired", Int. J. Res. in Eng., Sc. and Management, vol. 2, p. 4, 2019.

[15]

K.N. Kumar, P. Surendranath, and K. Shekar, Assistive device for blind, deaf and dumb people using raspberry-pi, 2017.https://www.onlinejournal.in/IJIRV3I6/048.pdf

[16]

J.M. Fletcher, W.A. Coulter, D.J. Reschly, and S. Vaughn, "Alternative approaches to the definition and identification of learning disabilities: Some questions and answers", Ann. Dyslexia, vol. 54, no. 2, pp. 304-331, 2004. [http://dx.doi.org/10.1007/s11881-004-0015-y] [PMID: 15741940]

[17]

R.M. Geiman, and W.L. Nolte, "An expert system for learning disability diagnosis", 1990 IEEE Int. Conf. Systems Eng, 1990pp. 363-366 [http://dx.doi.org/10.1109/ICSYSE.1990.203172]

[18]

K.P. Vinumol, A. Chowdhury, R. Kambam, and V. Muralidharan, "Augmented Reality Based Interactive Text Book: An Assistive Technology for Students with Learning Disability", 2013 XV Symp. on Virtual and Augmented Reality, 2013pp. 232-235 [http://dx.doi.org/10.1109/SVR.2013.26]

[19]

T. Adam, and A. Tatnall, "The value of using ICT in the education of school students with learning difficulties", Educ. Inf. Technol., vol. 22, no. 6, pp. 2711-2726, 2017. [http://dx.doi.org/10.1007/s10639-017-9605-2]

[20]

T. Adam, and A. Tatnall, "Use of ICT to assist students with learning difficulties: an actor-network analysis", In: Key Competencies in the Knowledge Society, N. Reynolds & M. Turcsanyi-Szabo, Eds., IFIP Advances in Information and Communication Technology. vol. 324. Germany, Springer, 2010, pp. 1-11. [http://dx.doi.org/10.1007/978-3-642-15378-5_1]

[21]

S. Luján-Mora, "Web accessibility among the countries of the European Union: a comparative study", Actual Problems Comput. Sci., vol. 1, no. 3, pp. 18-27, 2013.

[22]

M.A. Hersh, and S. Mouroutsou, "Learning technology and disability: overcoming barriers to inclusion: evidence from a multi-country study", Proc. IFAC-PapersOnline, vol. vol. 48, 2015pp. 8388 [http://dx.doi.org/10.1016/j.ifacol.2015.12.061]

[23]

Information Technology—W3C 2012.http://www.w3.org /WAI/

[24]

P. Acosta-Vargas, T. Acosta, and S. Luján-Mora, "Challenges to assess accessibility in higher education websites: a comparative study of Latin America universities", IEEE Access, vol. 6, pp. 36500-36508, 2018. [http://dx.doi.org/10.1109/ACCESS.2018.2848978]

[25]

W3C, Introduction to web accessibility..https://www.w3.org/WAI/fundamentals/accessibility-intro/

[26]

Publication Office, Accessibility to ICT products and services by disabled and elderly people..https://op.europa.eu/en/publication-detail/-/publication/

[27]

Recommendation for Information and Communication Technology (ICT) Accessibility Guideline for Persons with Disabilities..http://www.itu.int/themes/accessibility/dc/

[28]

J. Carter, and M. Markel, "Web accessibility for people with disabilities: an introduction for Web developers", IEEE Trans. Prof. Commun., vol. 44, no. 4, pp. 225-233, 2001. [http://dx.doi.org/10.1109/47.968105]

[29]

R.M. Bennefield, "Catching a view of the web", US News World Rep., pp. 68-69, 1997.

[30]

W3C, Fact Sheet for ‘Web Content Accessibility Guidelines 1.0., 1999.http://www.w3c.org/1999/05 /WCAG-REC-fact

Web

Content

Accessibility

Guidelines

(WCAG)

2.0.,

Accessibility

Advances in Data Science-Driven Technologies 47

[31]

C.D. Waddell, The growing digital divide in access for people with disabilities: Overcoming barriers to participation, 1999.http://www.aasa.dshs.wa.gov/access/waddell.htm

[32]

"IT Accessibility Guidelines: Application Software", Centre for Excellence in Universal Design, Ireland.http://universaldesign.ie/useandapply/ict/itaccessibilityguidelines/applicationsoftware/

[33]

M. Duplaga, "Digital divide among people with disabilities: Analysis of data from a nationwide study for determinants of Internet use and activities performed online", PLoS One, vol. 12, no. 6, p. e0179825, 2017. [http://dx.doi.org/10.1371/journal.pone.0179825] [PMID: 28662125]

[34]

J.N. Mottl, New Tools Boost Number of Disabled in IT Ranks, 2001p. 84.http://resna.stanford .edu/SIG-11/archive/boost.htm

[35]

M. Kulkarni, "Digital accessibility: Challenges and opportunities", IIMB Manag. Rev., vol. 31, no. 1, pp. 91-98, 2019. [http://dx.doi.org/10.1016/j.iimb.2018.05.009]

[36]

T. Noonan, Barriers to using Automatic Teller Machines: a review of the usability of self-service banking facilities for Australians with dissabilities.https://humanrights.gov.au/our-work/disabilityrights/barriers-using-automatic-teller-machines

[37]

P.A. Narayani, "Most ATMs remain inaccessible for the differently abled", https://www.the hindu.com/news/cities/Madurai/most-atms-remain-inaccessible-for-the-differently-abled/ article29668327.ece

[38]

D. Bouillet, and J. Kudek-Mirošević, "Students with disabilities and challenges in educational practice", Croatian J. of Edu., vol. 17, pp. 11-26, 2015. [http://dx.doi.org/10.15516/cje.v17i0.1472]

[39]

Mozilla Accessibility Project, Where Are We Today?.http://www.mozilla.org/access/today

[40]

BECTA, Information sheet on: visual impairment and ICT., 2001.Information sheet on: visual impairment and ICT, 2001.http://www.becta.org.uk/technology/infosheet

[41]

E. Bergman, and E. Johnson, "Towards accessible human interaction", In: Advances in HumanComputer Interaction. vol. Vol. 5. Ablex Publishing, 1995.

[42]

M. Trucano, Knowledge maps: ICTs in education. World Bank Group: Washington, DC, 2005.

[43]

R.A. Wildeman, and C. Nomdo, Implementation of inclusive education: How far are we?, 2007.http://www.idasa.org.za

[44]

H. Ahammed, "Challenges Faced by Teachers of Learners with Learning Disability", Int. J. Indian psychol., vol. 9, no. 2, pp. 294-312, 2021.

[45]

Access to telecommunication/ICT services by persons with disabilities and other persons with specific needs., 2021.https://www.itu.int/en/publications/ITU-D/pages/publications.aspx?parent=D-SG-SG01.07.5-2021&media=electronic

48

Advances in Data Science-Driven Technologies, 2023, 48-72

CHAPTER 3

Computer Vision-Based Assistive Technology for Blind and Visually Impaired People: A Deep Learning Approach Roopa G.M.1,*, Chetana Prakash1 and Pradeep N.1 Department of Computer Science and Engineering, Bapuji Institute of Engineering and Technology, Davangere – 577004, Visvesvaraya Technological University, Belagavi – 590 018 1

Abstract: According to the World Health Organization (WHO), a minimum of 2.2 billion individuals worldwide have impaired vision or are blind. In contrast to hereditary blindness, gained visual impairment is frequently identified as a result of aging, lifestyle habits, or hereditary influences. Aging-related presbyopia has the largest influence on visual impairment and is the second most prevalent cause of blindness globally, and the rate of acquired blindness is predicted to rise dramatically as life expectancy rises. When performing most of the everyday tasks that non-disabled individuals do, visually and blind people face several problems. Thus, assistive gadgets have been utilized to help the blind and visually impaired overcome physical, social, infrastructural, and accessibility hurdles to independence, allowing them to live engaged, creative, and fruitful life as equal members of society. The usage of assistance equipment has increased, and numerous electronic help devices have been produced in recent years, which have been superseded by traditional aid gear, such as white canes. Currently, ATs are created by integrating various types of sensors, cameras, or feedback channels that combine with various implementation methodologies to increase movement for the visually handicapped. Assistive systems based on computer vision or machine learning approaches have emerged, and as technology has advanced, so has assistive technology. Assistive technology is a priority in the field of education and rehabilitation for individuals with blindness or low vision because it “equalizes the ability to access, store, and retrieve information between sighted people and those with visual impairments”. Nowadays, technological advances are making a difference in their ability to overcome difficulties to some extent. Every day, they encounter a slew of challenges, the most significant of which are establishing one's position, determining one's heading and movement directions, and comprehending the placements of things. The goal of assistive technology is to boost impaired people's faith, comfort, security, independence, and quality of life by enhancing their mobility and decreasing their impairment.

Corresponding author Roopa G.M.: Department of Computer Science and Engineering, Bapuji Institute of Engineering and Technology, Davangere- 577004; E-mail: [email protected] *

Manoj Kumar M V, Immanuel Azaad Moonesar R.D., Ananth Rao, Pradeep N, Annappa, Sandeep Kautish and Vijayakumar Varadarajan (Eds.) All rights reserved-© 2023 Bentham Science Publishers

Deep Learning

Advances in Data Science-Driven Technologies 49

Keywords: Assistive Technologies, COCO-Dataset, Computer Vision, Object Recognition, OCR, TESSERACT, Visually Impaired, YOLO-v3.

INTRODUCTION The globe is transforming from machinery-employed or industrial economies to intelligence and information economizing in the twenty-first century. The goal of knowledge economy development is to address conventional societal hurdles, such as spatial variability of citizens, linguistic or knowledge barriers, handicaps caused by impairments or environmental circumstances, social position, and global influence. This might imply that the “digital divide” among developed/developing nations is diminishing. Will individuals with disabilities be included in the digital age resulting in considerable development and improvement in living circumstances in developing nations? The answer to the query will be “yes” when adequate attention is paid. The entry to the knowledgebased economy is Information and Communication Technology (ICTs), which consists of systems, conventional and smart cell devices, and laptops. Internet access is very important for disabled persons than the common public since they have fewer options for retrieving data or participating [1]. Currently, there are several methods in which ICTs are accessible and employed as assistive technology (ATs). On the other side, a critical element is the fast evolution of technology that acts as a tool and has a significant impact on how we acquire, play, and operate. This reliance on technology may be discretionary for most of us, but it is far less for individuals with disabilities, who frequently rely on technology to access jobs, complete everyday chores, and fully engage in the community; ICTs transforming the technology alternatives available to persons with impairments. Advanced materials also improve comfort and prevent skin deterioration in seating systems [2]. Wheelchairs and other mobility goods are becoming simpler to use and move in vehicles as materials get lighter. Changes in conventional home products (food preparation, self-care) are being created to suit elderly people who have arthritis, hearing problems, or vision loss. Notably, huge items, ranging from vehicles to household equipment with features, improve accessibility and usefulness for people with disabilities. As a result of such reasons, there are many more technological alternatives accessible to suit the requirements of disabled people. Some of them are popular items, while others are developed specifically for individuals with impairments. The function constellation of ICTs like mobile devices is similar to those necessary for various ATs related to mainstream techniques or custom-built devices. All that remains is to ensure that this occurs with new technology. One of

50 Advances in Data Science-Driven Technologies

G.M. et al.

the most precise technologies that can aid the blind incorrectly exploring their environment is the object detection system for the visually impaired [3]. This assists in recognizing the barrier and traveling from one location to another. Finally, it generates audio data about the item. Consequently, it would be easier to operate and portable for the visually handicapped. It also helps the blind understand their surroundings without the help of another person. In its broadest sense, the term “Assistive Technologies” (AT) relates to any grouping of technological advances (items, environmental modifications, facilities, and procedures) that may be utilized to address deficiencies and/or improve human function. AT, in particular, strives to assist persons with impairments or learning difficulties in coping with their daily surroundings and obtaining a greater quality of life. In general, AT is utilized in two major application scenarios: 1) health care, which attempts to alleviate (remediate) cognitive and behavioral deficits, and 2) social, which works on the surrounding community and focuses on social barriers and injustices. In the past few decades, there seems to be a tremendous increase in demand for novel technology that would enhance the quality of life, for example, for elderly people or individuals with different talents, as well as for those who have various disorders but wish to improve their comfort. Researchers from many areas have used their knowledge to build new technologies to satisfy the demands of various assistive device application settings. To summarize, at least one innovation is created every day throughout the world, and progress in the field of automated systems is quick and constant. Without this, the world would not have envisaged visually challenged individuals wearing a digital helmet integrated with a portable monitor, audio sensors, and stereo headphones to assist them in navigating. Thus, the breakthrough in wireless technology uses a sensor to take photos and process the acquired images to recognize things and give an audio message via stereo headphones. This may be useful in any indoor/outdoor setting. In this chapter, we examine the broad range of developing technical capacity as well as the particular significance of such technologies for persons with impairments. Few studies have noted that “although an advancement of technology and consumerism which is so concerning in other ways, many individuals remain excluded and handicapped by design which does not recognize their talents.” Inscribing this problem by gaining access to specialized and mainstream technology will allow persons with disabilities to engage fully in all society sections. Furthermore, a camera, a machine, and an audio output device comprise the main component based on computer vision for locating and

Deep Learning

Advances in Data Science-Driven Technologies 51

navigating help for blind people [4]. A mini-camera installed on a hat or sunglasses will gather visual data, while an image processor and voice output will be given by a computing device (with speech output via a Bluetooth earpiece). Auditory signals can be used to communicate the recognition output to blind users (speech/sound). The primary objective of this research is to assist visually impaired persons in detecting and identifying objects. Blind individuals would be able to utilize this technology to recognize the texts displayed on the objects as well. This experiment is intended to assess the YOLOv3 algorithm's capacity to find items, detect text, and provide audio output to aid a blind user in recognizing an object. The organization of the sections in the chapter includes the introduction, background theory exploring the drawbacks of the existing approaches, challenges, the methodology proposed, System Architecture with module description, details on the YOLO algorithm and COCO dataset, experimental setup, results and discussion, conclusion/future scope of the work and references. THE GLOBAL ASSISTIVE TECHNOLOGY COMMUNITY AND ITS IMPACTS ON PEOPLE WITH DISABILITIES Assistive technology (AT) is commonly defined as “any object, unit of machinery, or product/service, generally purchased commercially off the shelf, adapted/customized to develop, preserve, or enhance cognitive skills of persons with impairments.” Few changes are integrated into standard devices, where a majority of custom-made ATs satisfy the particular needs of disabilities. This constant stream of functionality has been going on for quite some time. Cassette recordings and rapid records were created to aid speaking books for the impaired. The typewriter was created for the impaired, the fountain pen for people who cannot pick a pen with a point owing to poorly skilled tasks, and the expert miter block for those who could not chop with both hands. Telephones were created to help the deaf and blind, but their effects were viewed largely in the broader society, effectively suppressing the deaf and hard-to-hear. However, the same technology that permitted telephony was also useful in detecting and magnifying sound in hearing aids. Furthermore, the revelation that phone lines could be utilized to send digital information via routers alleviated the issues of severely deaf persons [5]. This enabled the TTY to be used as a visual replacement for audible information transmitted through telephone lines. It also enabled the use of such SMS (texting) on mobile phones. Material and manufacturing processes are also included in the contributions to the general concept design, materials, and techniques, initially designed for individuals with

52 Advances in Data Science-Driven Technologies

G.M. et al.

disabilities. This “transition” of functionality from ATs to a model defined and again has aided a wider trend centered on universal design. PRESENT-DAY SCENARIO Around the years from 2013-2014, wearable devices became popular. This method refers to gadgets that may be worn on the wrist/neck. Fitbit, Pebble watches and Google glasses are notable features of wearable technology products. A technique called Finger-Eye comprises an electronically worn finger-like device with an integrated camera on the fingertips. The camera scans the text beneath the user's finger and reads it out to the blind user. However, the fingereye was not completed successfully and was only evaluated with a mounted camera. Few researchers have presented an event camera-based approach for selfdriving automobiles. Action cameras are a camera that captures events or changes in intensity values, which distinguishes their output from that of ordinary cameras. These action cameras function as motion sensors, detecting the motion of nearby objects and eliminating redundant data. GENERAL DESIGN IDEAS AND THE USABILITY OF DAILY ITEMS Donald Norman pioneered the notion of daily object design in the 1990s. Concerned by the difficulties he and others had faced employing everyday products that were badly built, he aimed to improve the architecture of the things that were encountered regularly. Norman noticed that the performance of many goods was neither apparent nor simple. He described this discrepancy between visible characteristics and functional activity using the notion of “affordances.” Affordances are defined as the “perceived and real characteristics of the item.” The word limitations were used by Norman to describe the restrictions on the number of alternative applications for an object. Norman's limitations fall into four categories: physical (based on actual-world qualities), semantic (based on significance), cultural (based on recognized customs), and logical [6]. The link between the spatial and physical arrangement of elements and the activities they govern is described in the latter category. Errors occur as an output of incorrect aim in mind, described as system affordances, behaviour impressions, and user device characteristics. Norman recommended linking affordances and restrictions be utilized to produce potential executions of every item to enhance the design. He also discussed perceptual models, which allow for predicting the influence of actions on an item. The sensory input that the object offers to the user is influenced by object perception. This feedback can be physical (for example, switch click or the force required to turn a knob), visual (for example, moving when triggered, lights or other indications), or aural (e.g., auditory alarms/signals).

Deep Learning

Advances in Data Science-Driven Technologies 53

EVOLUTION OF ASSISTIVE TECHNOLOGIES In recent years, several assistive technologies based on cameras, computer vision and IoT, have been released to aid blind people. Whereas such approaches have their own set of advantages and disadvantages. Handheld assistive gadgets, for example, rely on embedded systems like the Raspberry Pi. In these applications, computer vision is also used to detect logos and inscriptions on packed items. These gadgets are only useful for assisting blind individuals while shopping at a store. The battery backup of such devices, as well as the device's limited computing capacity, are other causes of worry. Mobile devices, which are more portable and include an integrated camera, are another option for aiding the blind. Few studies have suggested Suitable Parameter (QR) code-based systems for recognizing items. This technology is also connected with supermarket shopping and can recognize goods with QR codes. To make this gadget work, every product must be labelled with a QR code, which is impossible. Many works have presented a smartphone strategy for aiding blind individuals by identifying food types. As this version is based on the OCR algorithm, it inherits both the algorithm's benefits and shortcomings [7]. This software fails to recognize the product sign in the context of the size and illumination changes. Based on the shortcomings of current assistive offerings, there has been an advancement of assistive technologies that use current innovative technologies, and an effort was made to implement an efficient and robust object detection and recognition service (object detection/recognition with text-to-speech AT), which can assist people with intense impaired vision to autonomously navigate the world [8]. So far, numerous assistive techniques (vision substitutes) have been created, which may be classified as RFID-based methods, sensor methods, image processing ways, and computer vision approach. Amongst them, computer perception techniques for object recognition appear to be the most promising. It is witnessed that technological development in recent years, encourages us to create a system for the less privileged people. Numerous electronic assistance systems have been created so far, but relatively few of them use computer vision, while vision-based devices are gaining traction in the latest studies. In Fig. (1a) camera gadget is designed to be worn on the finger and aimed in the desired direction. In Fig. (1b and 1d ) are cameras placed on an eyewear system that is meant to function like an eye, allowing a person to turn his head in the direction of a possible object. In Fig. (1c and 1f ) a stereoscopic cane is displayed, which uses 3D imaging to obtain depth information. Fig. (1e) shows a smart cane with a laser vibratory component A few of these devices are connected for navigation, while others are designed for specific applications like banknote

54 Advances in Data Science-Driven Technologies

G.M. et al.

identification, as in Fig. (1d)

Fig. (1). Assistive Devices using Computer Vision

ASSISTIVE TECHNOLOGIES: FUNCTIONAL FRAMEWORK The variety of technological choices available to individuals with impairments for functional assistance is expanding. Major mainstream gadgets are now being utilized to give functional support that formerly needed specially developed assistive technology. However, specialists ATs are still necessary in many cases with a combination of technologies capable of meeting a wide range of requirements [9]. At one end of the spectrum are gadgets that give minimal help or increase a person's capacity to fulfill a job. For instance, a person with autism may be able to talk, but their speech can sometimes be hard to appreciate. In these cases, they may use a letter panel to spell out terms that are not comprehended. A person having respiratory issues, for example, can ambulate within their home but, due to limited stamina, may need a motorized wheelchair to conduct their shopping unassisted [10]. ATs that replicate substantial quantities of capacity to create functional results are at the opposite end of the spectrum. Some people, for example, have no vocal communication capacity and may need the gadget to communicate. Similarly, some people rely entirely on a conventional or power wheelchair for independent movement [11]. Hard-Soft Technologies Hard technologies are elements that are widely accessible and may be procured and integrated into AT devices. This ranges from basic mouth sticks to computers and software. AT device largely relates to hard technologies as defined by a few researchers. The fundamental trait that distinguishes hard technologies is their

Deep Learning

Advances in Data Science-Driven Technologies 55

tangibility. Soft technologies are related to human domains to draw decisions, initiatives, learning, idea creation, and service provision. Soft technologies may be classified into three types: humans, speech, and computers. These components of technology, without which hard technology never succeeds, are far more difficult to get since they rely heavily on human understanding rather than actual items [12]. The creation of successful usage methods also has a significant impact on the effectiveness of the AT method. Initially, the creation of these techniques may be highly reliant on the AT practitioner's expertise, experience, and inventiveness. With increasing expertise, the AT user develops methods that promote effective device usage [12]. In the current human/device interface, it is possible to assign certain tasks to the human, a few to the device, and others to personal assistance. We can assign certain tasks to humans, a few for technology, and others to a private assistant in any human/device system. Few studies describe various functional allocation techniques that are utilized in common human aspects of designing [13]. A few of these apply to design AT devices that help assess how and what sort of AT would be serviceable for individuals. The most basic method is comparative allotment which works to be completed and is entirely allocated to a human or a gadget in this case. The user's abilities determine the tasks that may be allocated, and the technology's features decide which abilities are allotted to it. A telephone, for example, is constructed with the presumption that the person can handle the device, ring up, perceive another person, and talk over the phone. All of such responsibilities have been delegated to the user. However, if the user is unable to complete any of these activities, the AT must give an alternate series of tasks. Assume, for example, that a certain customer can do all duties except hold phone dialing. Using a Bluetooth headset eliminates the need to clench the phone, and automated voice recollection may be used to input numbers and operate menus. The AT component of such a system is composed of these. When matching technological features to a consumer's talents, we frequently utilize comparative allocation. We employ this technique in AT devices wherever practical by combining the usage of the AT systems with Personal Assistance Services (PAS) [14]. The extent of the users and technological elements is not set; rather, it varies depending on the precise activities and tasks to be performed. Initially, a beginner mobile phone user may depend on instinct and use most often basic capabilities, such as dialling every number. More sophisticated capabilities, like contact lists, messaging, and others, can be employed as the understanding of the device's functionality grows, and methods are established. More duties, such as memorizing numbers, are allocated to the gadget in this manner, freeing the user to perform other things [15].

56 Advances in Data Science-Driven Technologies

G.M. et al.

OBJECT RECOGNITION Object detection is an integral component of a vision system that allows sustaining in this universe. Humans and other life forms can execute immediately and effortlessly, but this is a difficult problem for devices to solve since each object in the 3d environment can set an unlimited amount of 2D predictions owing to affine reconstruction, lighting changes, and camera angle. Object detection is a well-designed area with two main categories: simplistic learning and deep learning object identification procedures. The first technique leverages handdesigned attributes, such as SIFT, SURF, and HOG in the first phase, and the different stages are generally containers of quote and maze matching, with the later phases of such a channel being a classifier such as Classification Algorithm (SVM), K Nearest Neighbor (KNN), Artificial Neural network (ANN), and so on. While these approaches are extremely effective for specific applications, hand pattern descriptors are not as effective for generic item detection due to the significant diversity in form, texture, and so on. The most current technique for solving the object recognition challenge is to use Deep Learning-based methodologies, which involve designing a deep neural network and further training with huge instances in a supervised aspect, which consumes huge time to instruct but outperforms all hand-drawn techniques [16]. Deep learning model design may be taught in two directions: supervised learning with huge examples and unsupervised learning with small data. Unsupervised learning trains the network using unlabeled data, which requires an extremely small learning curve and does not require tagged input [17]. Convolutional neural network (CNN) and recursive neural network (RNN) are employed in a single stream. Gabor attribute extraction is headed by a pooling process that passes to the iterative neural network in the second stream. The merged attribute vector is employed to train the Softmax classifier to gain higher accuracy while requiring fewer features for faster operation [2]. Fig. (2) shows the general pipeline for object detection. While all these methods can be applied to classification or semantic segmentation tasks by simply averaging samples, this is not the case for object detection, where detection sample bounding boxes must be accurately associated and merged [10].

Fig. (2). Typical object recognition pipeline.

Deep Learning

Advances in Data Science-Driven Technologies 57

BACKGROUND THEORY In this part, we present related research on assistive technologies for people who are blind and object identification algorithms. We explore the algorithms’/systems’ capabilities as well as their disadvantages and limits. Object Detection Algorithms SIFT (Scale Invariant Feature Transform) Algorithm Ever since Deep Learning gained control, the SIFT method being one of the most used techniques for object identification. Few works have demonstrated various methods for object tracking and identification in images. The SIFT technique is used to extract important characteristics in the frame. Then, these important characteristics are clustered to recognize the moving object using an enhanced kmeans clustering method, which has employed log-polar transformation for stabilizing the video, making the video frame resistant to scale and rotation [18 21]. The results of the preceding tests demonstrate that SIFT has good precision but is sluggish. The results demonstrate that their technique generally operates well but suffers from standard item identification challenges such as scale, clutter, and variations in lighting. SURF (Speeded Up Robust Features) SIFT’s accuracy was good, and it could manage to scale. The downside was the algorithm’s pace, which was extremely sluggish. A new method, SURF, was developed to improve the efficiency of object identification. These researches cover SURF effectiveness advances as well as compare studies with approaches like SIFT, ORB, BRIEF, and others. The improvements are intended to make the SURF algorithm illumination stable and achieve greater matching frequencies. The following are the results of their experiment: ●

●

●

●

● ●

Using SURF in association with the optimal bin first algorithm leads to quicker matching. This could, however, improve matching quality. If geometrical mathematics is added to SURF, additional key points may be identified. Splitting the sampling units into two groups may decrease the descriptive dimensions. Illuminating variability may be accomplished by extracting descriptors from an image’s brightness order sequence. SURF over-performs SIFT in noisy pictures. SURF demands lower computational resources than SIFT.

58 Advances in Data Science-Driven Technologies

G.M. et al.

OCR(Optical-Character-Recognition) Another frequently used approach for object detection is OCR. It has mostly been used to identify text in images. The following tasks employ the OCR technique for item identification: Few observations have been conducted about using the OCR method for object detection [7, 22]. This study includes several tests, such as utilizing the OCR algorithm to identify food labels, expiration dates, and ID card details. All of these tests yielded the following results: ● ● ● ● ●

OCR operates well when interpreting data, with an efficiency of 70-90 percent. OCR suffers from fluctuating illumination and rotations. Reflective surfaces also cause problems for the OCR algorithm. OCR delivers correct information retrieval. Performance reduction in face detection instance.

Only a few studies used button detecting to assist robotic systems in traveling to their designated floor [23]. The authors merged OCR with Faster R-CNN to create OCR-RCNN, a single neural network. The testing results demonstrated that their method performs well even on untrained elevator panel pictures. YOLO (You Only Look Once) Since its inception, the YOLO (You Only Look Once) algorithm has gained popularity. The name implies glancing once to identify an object. Yolo exhibits a one-shot detector that supports the localization and classification of images in a single movement rather than two. This accelerates the entire process of object detection while introducing a small loss of accuracy. The YOLO algorithm has undergone two major upgrades over the years: Yolo v2/v3, with the initial method known as Yolo v1. Such techniques employ a pre-trained prototype produced by priming a dataset on a deep-learning model. The outcomes show that YOLO v3 is quicker and more accurate than its previous versions. ●

●

● ●

YOLO and Faster-RCNN exhibit comparable accuracy. Moreover, Yolo v3 outperforms R-CNN in terms of accuracy rate. Combining Spatial Hierarchy with YOLO can enhance the algorithm’s mean Average Precision (MAP). Increasing the detection scale can assist with identifying a smaller object. YOLO3 acquires some issues from YOLO v1, such as location issues when many small objects are close to one another.

Deep Learning

Advances in Data Science-Driven Technologies 59

R-CNN Some studies offer MaskLab as a solution, for instance, Segmentation issues in their work. MaskLab is constructed over Fast-RCNN, and attribute selection is made with ResNet-101. MaskLab comprises three key features: box recognition, feature extraction, and direction prediction. They evaluated MaskLab’s effectiveness using the COCO-based classification benchmark. MaskLab’s experimental examination yields some encouraging findings. These investigations involve offering an enhanced version to incorporate lighting awareness, optimizing regional proposals, and so on. They developed the system by utilizing a ResNet-101 model and Feature Pyramid N/W (FPN) [24]. The outcomes of these trials may be summed up as: ●

●

●

●

The experimental outcomes indicate that techniques could show the learning connections between various components. Optimizing the localized proposal production demonstrates efficient iterative refinements and outperforms previous RPN algorithms using AFW, WIDER, and Pascal Faces. Incorporating an illumination awareness network into Faster R-CNN increases its efficiency for pedestrian identification when compared to normal Faster RCNN. Using huge instances and basic heuristics programs enhances the algorithm’s efficiency.

Gaps Identified Some gaps that are addressed concerning object identification are: ● ●

Several situations lead to incorrect outcomes in identifying the objects. Today’s assistive devices can genuinely help blind people with special software, and using customized hardware may result in higher expenses.

Existing Assistance solutions for Blind People ●

Existing vision-based technologies assist visually challenged people in navigating their environment. This system is made up of a camera, a haptic response device that alerts impaired people for object detection, and an integrated system [23, 24]. This technology is worn by visual individuals, allowing them to move about. To find vacant spots along the travel path, regional suggestions are employed. The authors put their system to the test by

60 Advances in Data Science-Driven Technologies

●

G.M. et al.

having blind individuals go through a maze while wearing their gear [9]. The findings of these trials demonstrate that this technology substantially assists a blind user in navigating a track without colliding. However, navigating with this technique is slower than with a cane. The pace of navigation could be increased if this technology is paired with a walking stick for the visually impaired. Using the OCR technique, an assistive system is suggested to aid blind people in reading. This device takes the shape of a glove with camera-integrated indices that the blind user may wear. After putting on the glove, the impaired user would move their index finger from right to left over the first text. The camera beneath the finger executes the text pictures and provides audio output. The trial findings demonstrate how the finger can effectively read the text underneath the camera, however, the procedure is sluggish since the procedure creates a highquality image by combining numerous photos with similar frames [24]. The practicality is indeed a factor to consider, as the trials were carried out using a camera set on a table to read the text. If a fingertip camera is utilized, the real outcome may differ.

Few of the works suggested an assistive device to aid blind people; their method employs Raspberry Pi, an embedded computer used to process text in pictures. The camera is placed on the user’s spectacles and is linked to the Raspberry Pi. The experimental study demonstrates that the OCR system performs well when reading text, but performs poorly when detecting faces. With a fast detection rate, YOLO can achieve an accuracy of about 83%. Details of the Existing Assistance solutions for Blind People are given in Table 1. Table 1. Existing Assistance solutions for Blind People Algorithm

Advantages

Limitations

SIFT

High precision and robustness to scaling.

Requires high processing power, limited speed, poor performance with interference, fading, and changing light.

SURF

Maximum speed, robustness to transformations, and minimal computational power are all necessary.

Reduced accuracy over SIFT, finds few important points than other algorithms.

OCR

Text recognition is even more accurate.

The image prediction accuracy is poor, but it is not light-sensitive.

Regional Proposal Networks

Efficient, equipped to handle transitions, and adept at coping with lighting changes

The time needed to develop the model is rather long; it is a two-stage procedure that is slow than SSDs.

Deep Learning

Advances in Data Science-Driven Technologies 61

(Table 1) cont.....

Algorithm

Advantages

Limitations

Single Shot Detectors (YOLO)

It executes all computation in a single step, rendering it the quickest of all algorithms and resistant to modifications. Performs effectively in the presence of obstruction and variations in lighting.

Lower precision than SIFT

PRIMARY OBJECTIVE OF COMPUTER VISION Computer Vision comprises two objectives. From the standpoint of biological research, computer vision seeks to develop statistical methods of human vision. From an engineering standpoint, computer vision attempts to create autonomous devices capable of performing a few tasks that the visual system is capable of performing. Many vision challenges are linked to extracting 3D and periodic content from time-varying 2D data acquired with one or many news crews, as well as the general comprehension of such dynamic situations. The two objectives are inextricably linked [25]. Researchers creating computer vision solutions are frequently inspired by the features and characteristics of visual perception. In contrast, computer vision algorithms can provide information about where the human visual system functions. METHODOLOGY PROPOSED Input: IP webcam is utilized to send images to the model created at 60 fps, and an additional process is performed on alternative frames to increase performance. Dataset: Common objects are used to prepare the model in context dataset. Model: You Only Look Once (YOLO) is a model that has gone through several confusing transformations. Text-to-Speech: The identified items class categorization from every frame would be a message, such as cat. We will extract the image's object attributes and include the top/mid/bottom & left/center/right positions in the cat class prediction. To convert the string to vocal output, we utilize pysstx3. Tesseract: It is the most popular OCR engine, with a high-quality OCR corpus. OCR employs AI for text recognition and picture recognition. Tesseract- OCR detects patterns in pixels, letters, words, and phrases. Output: It is frame stream as a video referee after obtaining the bounding box coordinates for each item recognized in our frames. For every 30th frame (30 fps), a voice response is planned, e.g., bottom left cat – a cat was spotted in the view of

62 Advances in Data Science-Driven Technologies

G.M. et al.

the camera at the bottom left. The proposed methodology followed to carry out the work is given in Fig. (3).

Fig. (3). Proposed Methodology adopted.

YOLOV3 ARCHITECTURE You Only Look Once, Version 3 (YOLOv3) is an object identification system that detects particular items in films, live feeds, or pictures. It is a sophisticated convolutional neural network (CNN) for real-time object identification. Furthermore, it is well-known for its great accuracy and ability to operate in realtime or be utilized for real-world applications. The YOLO method “just looks once” at the source image, which means it only needs one forward propagating pass via the network to generate predictions. Fig. (4) shows the standard YOLOv3 framework.

Fig. (4). YOLOv3 Framework for Object Detection.

Deep Learning

Advances in Data Science-Driven Technologies 63

CNN's are classifier-based systems capable of processing incoming pictures as organized arrays of data and detecting patterns between them. YOLO has the benefit of being far quicker than some other networks while maintaining accuracy. It enables the model to examine the entire image at the testing phase, allowing its predictions to be informed by the image's globalized world. Regions are “scored” by YOLO and other deep neural network algorithms based on similarities to specified classifications [15, 26]. High-scoring areas are recorded as positive detections of the class with which they most strongly identify. The previous version was updated for incremental progress and is now known as YOLO v3. There have been numerous object-tracking algorithms for a long time, so the rivalry is mainly on how precise and rapidly things are recognized. YOLO v3 has everything required for real-time object recognition and object classification. The improved design has residue skip connections as well as - sampling. The most notable characteristic of v3 is that it detects objects at three distinct scales. YOLO is a deep convolution system that generates its output by employing a 1 x 1 kernel to a feature space. In YOLO v3, detection is accomplished using a 1 x 1 detecting kernel on feature vectors of various sizes at three distinct locations throughout the network. EXPERIMENTAL SETUP MS COCO (Microsoft Common Objects in Context) is a large-scale dataset for object recognition, segmentation, key-point identification, and labeling dataset. There are 328K pictures in the collection. Annotations: The dataset includes annotations for: ●

●

●

● ●

●

Bounding Box Predictions: Provides object score for every bounding box. The objectiveness score is predicted using regression analysis. Captioning: Image interpretations in natural language (refer to MS COCO Captions). Key-point Detection: Over 200,000 images and 250,000 human instances were tagged with 17 key points. Image extraction: Per-pixel separation masks for 91 different stuff categories. Panoptic: Complete scene classification with 80 item categories (such as human, bicycle, and elephant) and a subset of 91 thing categories (grass, sky, road). Dense Pose: Dense Pose annotations have been applied to over 39,000 pictures and 56,000 person samples; each identified person is labeled with a sample id and a mapping among image pixels belonging to that human body and a

64 Advances in Data Science-Driven Technologies

G.M. et al.

reference 3D model. Only the annotations for training and validation pictures are made public. The COCO dataset has been labeled, so it may be used to train supervised computer vision techniques to recognize similar objects in the dataset [20, 27]. These models are, of course, very far from perfect, therefore, the COCO dataset serves as a baseline for assessing the periodic development of these approaches through computer vision study. This project aims to achieve real-time recognition accuracy for blind individuals to make traveling and mobility easier for visually impaired people by utilizing handheld camera equipment. To assure the test's independence, the test participants were blinded at every trial and confused at the start of every trial. We used a head-mount for the mounting method of any compact camera equipment. As an experimental configuration, we utilize a Smartphone placed on the head using the Android app “IP Webcam.” The subjects that wore our system found it easy to transport because it is wireless and lightweight, and any tiny dimensions portable camera equipment may be put on it. Following the system's startup, the performance of the test subject was monitored to ensure that the items in the environment were properly detected and that the output help was adequate. The individual was then requested to match his knowledge with what the software was assisting him with, and the aided results were compared with the real necessary results to confirm the system's accuracy in detecting numerous items in the surroundings. It is also examined to see if the items identified are in the places suggested by the program. The patient recognized the various things without using his hands or any other physical help. Here's how to construct a basic head mount that will allow us to transmit video from the forehead to our smartphone. Materials and Tools: Cotton belt/strap, rubber bands, and hard cardboard, as shown in Fig. (5).

Fig. (5). Materials and tools required.

Deep Learning

Advances in Data Science-Driven Technologies 65

The supplies listed above are needed to create a small handmade experimental setup, which includes a cotton strap to wrap as a headband. In addition, cardboard and rubber bands are utilized to provide movable support, as shown in Fig. (6).

Fig. (6). Basic experimental setup.

The image above depicts the whole configuration of our portable device, which can be used with any portable camera equipment. This setup prototype is the basic form; it may be adjusted and extended further based on the demands and standards. The primary objective of this research is to assist visually impaired persons in detecting and identifying objects. Blind individuals would be able to utilize our technology to recognize the texts displayed on the objects as well. This experiment is intended to assess the YOLOv3 algorithm's capacity to find items, detect text, and provide audio output to aid a blind user in recognizing an object. RESULTS AND DISCUSSION The suggested framework is made up of two components: object detection and OCR. Object detection's primary goal is to evaluate the existence of items in the view in front of people, whereas OCR delivers text to users. For the gross orientation challenge, the system performed quite well. Objects within several meters of the observer were identified efficiently. As every 30th frame is collected and analyzed, real-time processing with a continual change in location was occasionally complex but typically acceptable. It has been noted that the vocal output for the identified items was straightforward and intuitive to comprehend and that it is successful since it does not cause an overload of information and consumes less effort to get used to it. Following that, the simplicity of usage was enhanced. During OCR detection, the system identified the text extremely rapidly, and it was subsequently transformed into spoken output with quite a high accuracy, including local languages.

66 Advances in Data Science-Driven Technologies

G.M. et al.

System Work-Flow for Object Detection Fig. (7a and 7b) shows the system workflow and the object detection module adopted. IP Camera feeds real-time video to the server

The server detects the objects in the environment with their spatial locations

Image Capturing System

Video Frames

YOLOv3

Keypoint Description The text is read aloud using TTS engine

The outputs are saved as text in real time

Keypoint Detection

Image Preprocessing Registered Features Extracted From Objects

Descriptor Matching

Speech System

Fig. (7). (a). System workflow diagram. (b). Object Detection Module

Fig. (8). A person and cell phone is detected with their correct spatial positions

Fig. (9). Table and chair are detected with their correct spatial positions.

Frame(t)

Deep Learning

Advances in Data Science-Driven Technologies 67

Fig. (8 and 9) shows the details of the object detected with their correct spatial positions SMART READING SYSTEM FOR VISUALLY IMPAIRED PEOPLE USING TESSERACT We present a smart device that helps the visually impaired read paper-printed text. A camera is used as a device that can be used for reading text documents. Based on studies with Blind people, the design is made portable. The proposed system feeds data into the system using a portable IP webcam which is then processed by Tesseract [6, 28]. The OCR software and the Text-to-Speech (TTS) are the fundamental blocks used as the basis for most access technology solutions designed for people with blindness and reduced vision. Optical character recognition (OCR) is the translation into the machine-encoded text of the recorded images from printed text. OCR is a mechanism in which objects (letters, symbols, and numbers) relate a symbolic value to a character’s image. Optical character recognition is also beneficial for people who are unable to read a text document but intend to know its content. OCR enables the use of machine translation, textto-speech, and data extraction techniques in recorded or scanned documents. The final accepted text document is fed to the output system or a speaker that can read out text aloud. FLOW PROCESS OF TESSERACT ●

●

●

●

●

Image Capturing: It is the step where the cell phone captures the image with text on it. To provide quick and consistent identification through the high-resolution camera, the image quality should be high. Pre-processing: Noise is removed in the pre-processing stage. The image is scanned for skewing. Skewing can occur either in the right or left orientation. The image is brightened before moving to further processing. Segmentation: The picture is then transferred to the segmentation stage after pre-processing. In this process, it attempts to break down a picture of a series of symbols into the individual symbol sub-image. The picture histogram helps to measure the horizontal line width. The width of the words is detected using histograms. They are then broken down into symbols that use symbol width calculation. Feature Extraction: Feature extraction is the process of mining out features from a picture that are defined by character, height, and width, the horizontal and vertical lines, pixels in the various regions, etc. Image to Text Converter: The ASCII values are processed for known characters. Here each character matches its equivalent pattern and is saved as a regularized transcript of the text.

68 Advances in Data Science-Driven Technologies

G.M. et al.

The fundamental block of this system is Tesseract and gets; below (Fig. 11) demonstrates the workflow of the proposed system.

Fig. (10). (a). System work-flow of OCR (b). Text Recognition System.

Fig. (11). The text on the screen is detected with an accurate output.

Fig. (12). The text on the screen is detected with an accurate output.

Deep Learning

Advances in Data Science-Driven Technologies 69

Fig. (13). The text on the screen is detected with an accurate output.

The flow diagram of the OCR working in an experimental setup is shown in Fig. (10a and 10b) where the live video captures the text images as input which is then fed to the server for processing, and the subject is assisted with audio output. Fig. (11-13) shows the accurate output drawn by detecting the text on the screen. FUTURE RESEARCH DIRECTIONS Despite recent advances in object recognition, some issues demand special consideration. ●

●

●

●

● ●

From the most recent developing trends, such as brain-inspired computation and collaborative learning. The benefit and advantage of object recognition may be increased or employed in other modalities, such as gesture recognition. Particular components of object recognition are rudimentary and require substantial improvements, such as tagged training sets and large amounts of data to train. The computational tool proved efficient and successful in a specific sector while also developed as a powerful generalized tool. Lacks a basic detector that meets the complete criteria. An innovative method for optimizing hyper-parameters for deep learning modules.

CONCLUSION Almost every element of our daily life is influenced by technology. Only a few tasks are performed without the aid of technology. AT is generally conceptualized as technology designed specifically for persons with disabilities to augment or

70 Advances in Data Science-Driven Technologies

G.M. et al.

replace function. This word is frequently used to describe the devices required by several pieces of policy that aid in the accessibility and inclusiveness of individuals with disabilities. However, in recent decades, technology developed for common use (i.e., currently accessible and not specifically created for persons with disabilities) has proven useful to people with a wide range of abilities. Information, communication, and computing infrastructure, particularly, contain features that allow them access to consumers of varying capacities. YOLOv3 is a revolutionary deep learning-based system for object identification and categorization. Our suggested application captures images in real-time and sends them to the obstacle recognition system. The experiment results demonstrate the deep efficacy of the system, which is capable of not only displaying the output of the obstacle identified and classifying the obstacle but also generating audio output in their native languages. The use of obstacle detection and categorization for visually impaired individuals would improve their safety and comfort, resulting in a higher quality of life daily. We plan to investigate the distance between visually impaired persons and barriers in the future. We intend to investigate a comparable triangle, Euclidean distance, and other hypotheses before incorporating them to enhance the application. CONSENT OF PUBLICATION The Figures used in this chapter are in the public domain (copyright-free) and are under other open licenses. Hence, no consent is to be taken from anyone for the materials used in this chapter. CONFLICT OF INTEREST The authors have no Conflict of Interest to declare. All co-authors have seen and agree with the contents of the manuscript, and there is no financial interest to report. ACKNOWLEDGMENT The authors of the chapter would like to acknowledge our colleagues who initially reviewed the chapter before submitting it to the editors. Also, the authors are thankful for the non-teaching staff who stayed with us after working hours while documenting and fine-tuning the chapter. REFERENCES [1]

J. Deng, X. Xuan, W. Wang, Z. Li, H. Yao, and Z. Wang, "A review of research on object detection based on deep learning", In: The 2020 International Seminar on Artificial Intelligence, Networking and Information Technology.Shanghai, China., 2020.

[2]

L. Liu, W. Ouyang, X. Wang, P. Fieguth, J. Chen, X. Liu, and M. Pietikäinen, "Deep Learning for Generic Object Detection: A Survey", Int. J. Comput. Vis., vol. 128, no. 2, pp. 261-318, 2020.

Deep Learning

Advances in Data Science-Driven Technologies 71

[http://dx.doi.org/10.1007/s11263-019-01247-4] [3]

X. Wua, "DoyenSahoob, Steven C.H.Hoiab “Recent advances in deep learning for object detection”", Neurocomputing, vol. 243, no. 21, pp. 125-132, 2020.

[4]

C.Y. Cao, J.C. Zheng, Y.Q. Huang, J. Liu, and C.F. Yang, "Investigation of a Promoted You Only Look Once Algorithm and Its Application in Traffic Flow Monitoring", Appl. Sci. (Basel), vol. 9, no. 17, p. 3619, 2019. [http://dx.doi.org/10.3390/app9173619]

[5]

J. Hui, What do we learn from single-shot object detectors (ssd, yolov3), fpn & focal loss (retinanet)?, vol. 28, 2019.

[6]

Z. Zhang, T. He, H. Zhang, Z. Zhang, J. Xie, and M Li, "Bag of Freebies for Training Object Detection Neural Networks", arXiv:1902.04103 [cs], 2019.

[7]

Y. Tian, G. Yang, Z. Wang, H. Wang, E. Li, and Z. Liang, "Apple detection during different growth stages in orchards using the improved YOLO-V3 model", Comput. Electron. Agric., vol. 157, pp. 417426, 2019. [http://dx.doi.org/10.1016/j.compag.2019.01.012]

[8]

M. Najibi, B. Singh, and L.S. Davis, "Fa-rpn: Floating region proposals for face detection", The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Jun 10, 2019. [http://dx.doi.org/10.1109/CVPR.2019.00791]

[9]

P. Ammirato, and A.C Berg, "A Mask-RCNN Baseline for Probabilistic Object Detection", arXiv:1908.03621 [cs], 2019.

[10]

D. Miller, F. Dayoub, M. Milford, and N. Sunderhauf, "Eval- ¨ uating Merging Strategies for Sampling-based Uncertainty Techniques in Object Detection", IEEE International Conference on Robotics and Automation (ICRA), 2019.

[11]

C. Li, D. Song, R. Tong, and M. Tang, "Illumination-aware faster R-CNN for robust multispectral pedestrian detection", Pattern Recognit., vol. 85, pp. 161-171, 2019. https://linkinghub.elsevier.com/ retrieve/pii/S0031320318303030 [http://dx.doi.org/10.1016/j.patcog.2018.08.005]

[12]

M. Sharif, S. Khan, T. Saba, M. Raza, and A. Rehman, "Improved video stabilization using sift-log polar technique for unmanned aerial vehicles", International Conference on Computer and Information Sciences (ICCIS), pp. 1-7, 2019. [http://dx.doi.org/10.1109/ICCISci.2019.8716427]

[13]

B. Benjdira, T. Khursheed, A. Koubaa, A. Ammar, and K. Ouni, Car Detection using Unmanned Aerial Vehicles: Comparison between Faster R-CNN and YOLOv3. IEEE, 2019, pp. 1-6.

[14]

A.I. Maqueda, A. Loquercio, G. Gallego, N. Garc’ıa, and D. Scaramuzza, "Event-based vision meets deep learning on steering prediction for self-driving cars", 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 5419-5427, 2018. [http://dx.doi.org/10.1109/CVPR.2018.00568]

[15]

A. Wong, M.J. Shafiee, F. Li, and B Chwyl, "Tiny SSD: A Tiny Single-shot Detection Deep Convolutional Neural Network for Real-time Embedded Object Detection", arXiv:1802.06488, 2018

[16]

S.A.K. Tareen, and Z. Saleem, A comparative analysis of SIFT, SURF, KAZE, AKAZE, ORB, and BRISK, 2018.https://ieeexplore.ieee.org/document/8346440/ [http://dx.doi.org/10.1109/ICOMET.2018.8346440]

[17]

L.-C. Chen, A. Hermans, G. Papandreou, F. Schroff, P. Wang, and H. Adam, Mask-Lab: Instance Segmentation by Refining Object Detection with Semantic and Direction Features, pp. 4013-4022, 2018.

[18]

P. Tang, X. Wang, A. Wang, Y. Yan, W. Liu, J. Huang, and A. Yuille, Weakly Supervised Region Proposal Network and Object Detection, vol. 11215, pp. 370-386, 2018.

72 Advances in Data Science-Driven Technologies

G.M. et al.

[19]

P. Maolanon, and K. Sukvichai, Development of a Wearable Household Objects Finder and Localizer Device Using CNNs on Raspberry Pi, vol. 3, pp. 25-28, 2018.

[20]

H. Nakahara, H. Yonekawa, T. Fujii, and S. Sato, A Lightweight YOLOv2: A Binarized CNN with A Parallel Support Vector Regression for an FPGA, pp. 31-40, 2018. [http://dx.doi.org/10.1145/3174243.3174266]

[21]

X. Zhang, W. Yang, X. Tang, and J. Liu, "A Fast Learning Method for Accurate and Robust Lane Detection Using Two-Stage Feature Extraction with YOLO v3", Sensors (Basel), vol. 18, no. 12, p. 4308, 2018. [http://dx.doi.org/10.3390/s18124308] [PMID: 30563274]

[22]

P. Deekshitha, B.T. Reddy, S. Badadha, K. Dhruthi, and S. Pandiaraj, "Object motion perception and tracking using sift with k-means clustering", Journal of Network Communications and Emerging Technologies, vol. 8, no. 4, 2018.www. jncet. org [JNCET].

[23]

R. Widyastuti, and C-K. Yang, Cat’s Nose Recognition Using You Only Look Once (Yolo) and ScaleInvariant Feature Transform. SIFT, 2018, pp. 55-56.

[24]

K.-J. Kim, P.-K. Kim, Y.-S. Chung, and D.-H. Choi, Performance Enhancement of YOLOv3 by Adding Prediction Layers with Spatial Pyramid Pooling for Vehicle Detection., pp. 1-6, 2018. [http://dx.doi.org/10.1109/AVSS.2018.8639438]

[25]

M. Rajesh, B.K. Rajan, A. Roy, K.A. Thomas, A. Thomas, T.B. Tharakan, and C. Dinesh, Text recognition and face detection aid for visually impaired person using Raspberry PI, 2017.http://ieeexplore.ieee.org/document/8074355/ [http://dx.doi.org/10.1109/ICCPCT.2017.8074355]

[26]

H-C. Wang, R.K. Katzschmann, S. Teng, B. Araki, L. Giarre, and D. Rus, Enabling independent navigation for visually impaired people through a wearable vision-based feedback system, 2017.http://ieeexplore.ieee.org/document/7989772/ [http://dx.doi.org/10.1109/ICRA.2017.7989772]

[27]

Y. Kim, and H. Jung, Re-configurable hardware architecture for faster descriptor extraction in SURF, 2017.https://digital-library.theiet.org/content/journals/10.1049/el

[28]

Z.X. Geng, and Y.Q. Qiao, An Improved Illumination Invariant SURF Image Feature Descriptor, 2017.https://ieeexplore.ieee.org/document/8719146/ [http://dx.doi.org/10.1109/ICVRV.2017.00090]

Advances in Data Science-Driven Technologies, 2023, 73-97

73

CHAPTER 4

Assistive Technology for Home Comfort and Care Annu Rani1,*, Vishal Goyal1 and Lalit Goyal2 1 2

Department of Computer Science, Punjabi University Patiala (Punjab), India Department of Computer Science, DAV College, Jalandhar (Punjab), India Abstract: Every second, individuals with physical and cognitive disabilities struggle so much to do some actions that normal people easily do within seconds. Assistive Technologies (AT) are those modules or sets of arrangements that aim to make life easy for disabled people, by stopping blockage and improving their mental and physical power. They improve their working capability, confidence, standard of living, and optimism. In modern times, Artificial Intelligence (AI) and technologies are developing rapidly, and new machines, motors, and mostly electronic devices powered by powerful batteries are being built every second. These are making it possible for disabled people to become self-dependent. Today, Assistive technology devices are efficient and suitable for disabled people. This chapter aims to provide in-depth knowledge about various types of disabilities, how disabled people face different problems and challenges, and how they can select and use assistive devices and mobile apps to live independently and comfortably.

Keywords: Ability, Activities, Aids, Assistive Technology, Comfort, Communication, Devices, Disability, Disabled People, Disease, Difficulty, Guardian, Help, Independence, Home, Individual, Mobility, Obstacles, Product, Person, Services. INTRODUCTION Assistive technologies are devices used to support the health and every activity of a disabled person [1]. It promotes the ability of a disabled person to perform activities of daily living (ADLs) independently. ADLs are self-care activities, such as eating, bathing, dressing, toileting, mobility, and personal device care. Assistive Technologies help to perform major and daily life activities that are otherwise impossible for the individual to carry out. The great principle of promoting ability includes a higher level of independence, reduction of spending * Corresponding author Annu Rani: Department of Computer Science, Punjabi University Patiala (Punjab), India Mob. 94787-75953; Tel: 94787-75953; E-mail:[email protected]

Manoj Kumar M V, Immanuel Azaad Moonesar R.D., Ananth Rao, Pradeep N, Annappa, Sandeep Kautish and Vijayakumar Varadarajan (Eds.) All rights reserved-© 2023 Bentham Science Publishers

74 Advances in Data Science-Driven Technologies

Rani et al.

time in daily living activities, and greater satisfaction in participating in daily activities. For example, wheelchairs provide great help in independent mobility for those who cannot walk, assistive eating devices can enable people who cannot eat food themselves, and hearing aids are useful for hearing-disabled people to hear more clearly, etc [2]. By using assistive technology, a disabled person has an opportunity for a positive and independent lifestyle and increased participation in social activities. This technology has proved a boon for disabled people by providing entertainment, security, a comfort zone, independent life, etc. Today, a lot of disabled people have gained high fame and name in the world by using AT. Several companies, such as Apple, Facebook, Flipkart, Google, Tata Group of Industries, and IBM Corporation, have increasingly focused on making their services more accessible for disabled people and building adaptive devices to improve their user’s quality of life. By current estimations, more than 4,000 assistive technology devices have been invented for elders and disabled people. This equipment includes everything from wheelchairs to a wide collection of high-technology devices, and many firms today are turning their focus and research to assistive technologies. There is an urgent requirement to consult a health care provider, such as a doctor, pharmacy technician, psychologist, or physical therapist to find out what is best available to fit the requirements. Various public and privates sources, such as UCP Bellow Funds, American Council of Blinds, Muscular Dystrophy Association, US Department of Veterans Affairs, National Multiple Sclerosis Society, etc. provide funds for AT devices. The organized structure of the chapter is as follows: Sect. 2 presents the disability and types of disabilities, Sect. 3 presents the common barriers faced by people with disabilities, Sect. 4 presents the principles for providing assistive devices, Sect. 5 presents various types of assistive technologies, Sect. 6 presents different types of mobile apps for disabled people, Sect. 7 presents the benefits of assistive devices in individual life and Sect. 8 includes a conclusion. DISABILITY More than one billion (1,000,000,000) people worldwide, making up approximately 15% of the total population, live with some kind of disability [3]. According to the WHO disability report, the population of disabled people is increasing daily because of the global rise in severe health conditions related to disability, such as hypertension, diabetes, respiratory illnesses, heart diseases, and mental disorder. Other environmental elements, e.g., floods, volcanoes, earthquakes, inaccessible transportation, road accidents and quarrels, are also responsible for increased disabilities.

Technology

Advances in Data Science-Driven Technologies 75

Disabled people often face discrimination in recruitment, salary, promotion, work, healthcare services, and educational institutions [4]. Usually, organizations and governments often overlook the needs of disabled people, even though people with disabilities are among the weakest, with lower educational success, lower job opportunities, and the poorest people. Poverty can be reduced if we include and take disabled people together. Types of Disabilities According to the RPWD (Rights of Persons with Disabilities) Act 2016, the list of disabilities has included a total of 21 disabilities, and all the disabilities are explained below. Blindness Blindness refers to a person with a blindness disability who is altogether unable to see by both eyes or sightless at all [5, 6]. Low Vision Low vision refers to the vision loss caused by a disease of the eyes, which cannot be improved or corrected via regular glasses, surgery, pharmaceuticals, or contact lenses. Hearing Disability This disability includes those people who are partially or completely deaf. They often use hearing aids and sign language for interaction with normal people. Hearing aids assist the hard of hearing people to hear sound clearly. Sign Language is a visual gesture language used by deaf people to convey what they want to say. Dwarfism Dwarfism or short stature, is a growth disorder characterized by a smaller height than usual. Intellectual Disability Intellectual disability, also called Mental Retardation (MR) or learning disability, is characterized by below-average brain functioning (decision-making, learning, thinking, and reasoning) and a lack of skills required for everyday life activities [7]. A person with an intellectual disability can learn new activities, but he/she learns slowly compared to normal people [8].

76 Advances in Data Science-Driven Technologies

Rani et al.

Autism Spectrum Disorder (ASD) ASD is a developmental and neurological disorder that causes troubles in social communication and behavior. The developmental disorder can be diagnosed in early childhood. Autism largely affects non-verbal communication, behavior, cognition, social interaction, speaking skills, physical health, and feelings of the affected person. Mental Illness Mental health disorder or mental illness means a substantial disorder of memory, thinking, spirit, mood, observation, orientation, or perception that grossly impairs judgment, emotions, manners, power to recognize real life, or capacity to meet the normal needs of life [9]. Locomotor Disability A person with a locomotor disability faces difficulty in mobility or walking from one place to another. Usually, it is considered a disability associated with legs, feet, limbs, injuries of the spine, and joints. It causes difficulties in individual’s movements, e.g., Holding objects in hands, raising both arms altogether or walking [10]. Leprosy Cured Persons Leprosy is an infective disease, also called Hansen’s disease (HD), caused by a slow-growing bacterium called Mycobacterium leprae. People with Leprosy suffer from loss of sensation in feet and arms, skin sores, disfigureuring lumps, peripheral nerve damage, loss of eyelashes, and pain in the joints [11]. Muscular Dystrophy (MD) Muscular dystrophy is a set of abnormal genes or inherited diseases that cause muscles to become weaker and lose muscle strength. Chronic Neurological Conditions Neurological diseases include a group of diseases such as Parkinson’s disease, Huntington’s disease, headaches, Stroke, cognitive disease, and Amyotrophic Lateral Sclerosis [11].

Technology

Advances in Data Science-Driven Technologies 77

Specific Learning Disability Specific Learning Disabilities include a collection of disabling conditions that curb a person's ability to reason, identify the character, learn, think, read, write, logic, listen, and correct spelling. Multiple Sclerosis(MS) MS is a potentially disabling disease that attacks the spinal cord and brain that cause interaction troubles between the brain and the rest of the body. MS symptoms include blurred vision, thinking problems, bone stiffness, weakness, fatigue, fuzzy memory, etc. Speech and Language Disability Speech and language disorder affects speech and learning ability due to various causes, such as brain injury, cognitive disorder, neurological disorder, hearing loss [12], etc. Thalassemia Thalassemia is a hereditary blood disorder in which the body produces abnormal or less hemoglobin. It is the responsibility of hemoglobin to carry oxygen from the lungs to all parts of the body. Consequently, there is a decrease in the excessive number of red blood cells, which leads to a new disease called anemia. Hemophilia Hemophilia is a blood disease due to a lack of clotting factors in the blood. A person with hemophilia may bleed for a longer period after a wound or injury than normal [13]. Sickle Cell Disease Sickle cell disease is a genetic set of blood disorders due to RBC (Red Blood Cell) breakdown. The main symptoms of sickle cell disease are pain, delayed growth, swelling of feet, dizziness, fatigue and infections.

78 Advances in Data Science-Driven Technologies

Rani et al.

Multiple Disabilities, including Deaf-Blindness Multiple disabilities combine two or more disabilities that affect listening, learning, visual and thinking power. Deafblindness is a multiple disability, the combination of both hearing impairment and visual impairment [14]. A person with a deaf-blindness disability faces extreme challenges in his life. Acid Attack An acid attack is a kind of cruelty or violence in which corrosive or acid is thrown at girls or women to kill them [15]. Parkinson’s disease (PD) PD is a brain disease that generates difficulty in movements, balance, fatigue and sleeping problems. Cerebral Palsy (CP) CP is the most common disability that occurs in children before birth because of abnormal brain development. CP disability affects the child’s ability to move and maintain balance and body posture. Types of disabilities are shown in below Fig. (1)

Fig. (1). Different types of disabilities

Technology

Advances in Data Science-Driven Technologies 79

COMMON BARRIERS FACED BY PEOPLE WITH DISABILITIES Every human being faces problems, tensions, and hardships at one time or another time. But, people with disabilities have more frequent and challenging hardships than healthy humans. Often there are various obstacles in the life of the disabled person, making it difficult or impossible for him to function. Communication Problem Communication problems are faced by those who have disabilities in speaking, hearing, understanding, reading, or writing, and they use different approaches to interact with normal people. Examples of communication obstacles consist of the following: ●

Audible messages may not be accessible for those people who have hearing disabilities consisting. Videos are playing without subtitles or captions. Verbal interactions without associated manual interpretation (such as British Sign Language). Normal people want to interact with Deaf people. But both of them do not know the other language. Deaf people cannot understand normal people’s language, and normal people do not know sign language. Here, the language is a communication barrier between both communities. Written messages may not be accessible for those people who have vision disabilities consisting. Use a small font size of print material without proper alignment. Braille is not available for the community that uses a screen reader. People with cognitive disabilities may not understand long sentences, technical words and some words with many syllables. ❍ ❍

❍

●

❍ ❍

●

Physical obstacles Physical obstacles are structural barriers in artificial or natural environments that stop or obstruct mobility. There are some examples of physical obstacles consist of: ●

●

Curbs and Stairs prevent an individual with a mobility disability from entering a shop, educational institute, hospital, restaurant, bank, etc. Not sufficient space around toilet rooms and other areas makes it difficult for a

80 Advances in Data Science-Driven Technologies

●

●

Rani et al.

wheelchair person to turn around. Mobility impairment students also faced the trouble of congested classrooms, lack of seats in the lab, classroom, etc [16]. Heavy doors will be hard for people with a mobility disabilities to open.

Social Obstacles ●

● ●

●

Social obstacles are related to the situations in which human beings are born, grow, understand, observe, feel, think, live, and operate– or social determinants of health care– that can put up to reduced working power among disabled people. People with disabilities get fewer jobs than normal people [17]. A few students with disabilities get a high school education compared to students without disabilities. Employees with disabilities get lower incomes as compared to healthy people.

Attitudinal barriers Attitudinal obstacles are the most fundamental and put up to other problems. For example, some people probably do not realize that troubles in getting on time or into a place can stop or limit a disabled person from participating in daily life activities. Some examples of attitudinal obstacles consist of: ●

●

●

Misconception: In society, normal people have wrong beliefs about disabled people. For example, normal people assume disability is a personal tragedy, a curse of God, and wrong deeds in the previous birth, etc. These beliefs make disabled people alone in society. Stereotyping: People sometimes think about disabled people, they are unfit, and the quality of their life is poor because of their disabilities. Discrimination and favoritism: People sometimes assume that people with disabilities are not able to do good quality work or are unable to participate in the decision-making process because of their disabilities. That’s why people with impairment get less employment and salary compared to healthy people.

Transportation obstacles Transportation obstacles are due to insufficient transportation that obstructs a person’s capability to be self-determinant and to operate freely in society. Some examples of transportation obstacles consist of:

Technology ●

● ●

●

Advances in Data Science-Driven Technologies 81

Lack of convenient transportation for those who cannot drive due to cognitive, vision or physical impairments. Unavailibity of public transportation, perhaps at unsuitable locations. Overcrowded public transportation, limited reserved seats and sometimes reserved seats already occupied by healthy persons. The conductors of the vehicle have no time for the person with disabilities.

PRINCIPLES FOR PROVIDING ASSISTIVE DEVICES The main goal of assistive devices and technologies is to maintain or improve an individual functionality and performance and help them live independently in society. A variety of devices are available in the market for disabled people. For example, vision aids, hearing aids, wheelchairs, and special computer software improve visual, hearing, mobility, and learning capacities. Availability Products and services are within easy reach in adequate amounts as close as possible to infants' or children’s communities. The children can obtain more benefits by using adaptive devices and live a more enjoyable life. Accessibility Products and services are reachable to everyone who requires them. Their supply or delivery should be without discrimination of color, age, religion, impairment groups, gender, creed and socio-economic groups. Physical accessibility means physical access to products and information; for example, easy entry into buildings, sufficient space to move around for wheelchairs, lights are appropriate, symbols presented in Braille format, seat availability, electronic switches, translating devices etc. Cognitive accessibility means that audible and written documents and instructions are in simple and unambiguous language, signs are presented in concrete format rather than abstract, services are easily available, and products are user-friendly—all from the perspective of disabled people [18]. Affordability Products and services are reasonably priced for the family of every person who requires them. Many of them will not be able to buy assistive technology devices unless it is delivered without any charge or subsidies.

82 Advances in Data Science-Driven Technologies

Rani et al.

Adaptability Products and services are modified and changed according to the demands and requirements of the person. They must adapt to differences in the condition of environmental elements (for example, physical, environment, cognitive environment, and society) as well as individual facets such as health condition, body posture, body capacity, intellectual level, age, and habits. Acceptability Products and services are suitable for every person. Factors such as efficiency, affordability, reliability, robustness, user-friendly, simplicity, protection, comfort and aesthetics should be considered to ensure that related services and equipment are suitable to the individual. In addition, available product designs, equipment materials, and colors should satisfy both girls and boys. Quality Services and products are well fit for use by an individual. Product quality can be estimated using technical standards or instructions in terms of safety, price, design, preference, shape, strength, user satisfaction, durability, and comfort. ASSISTIVE TECHNOLOGIES FOR HOME RELAXATION AND CARE FOR DISABLED PEOPLE In modern times, different assistive devices are available online, in the shop, malls, and markets. According to the requirement, disabled people select assistive aid to live an independent and comfortable life. Mobility aids A person who has a physical disability may require support to walk. Mobility devices targeted to assist with mobility aids include manual and powered wheelchairs, canes, crutches, walkers, scooters, prosthetics, artificial legs, supportive seats, corner chairs, walking sticks, tricycles, walking frames and orthotic aids [19]. These devices assist the user in various ways, including selfesteem, increased self-confidence and independence [20]. Different types of mobility devices are shown in below Fig. (2)

Technology

Advances in Data Science-Driven Technologies 83

Fig. (2). Mobility devices for disabled people [21].

●

●

●

●

Wheelchair: Wheelchair is a chair used by those people who cannot walk due to spinal cord injuries, cerebral palsy, muscular dystrophy, disability, illness, and problems related to old age, etc. Wheelchairs are also more comfortable for people with serious impairments or when traveling a long distance is required. Wheelchairs come in various designs, or formats to meet the specific requirements of their users. Examples of specific kinds of wheelchairs include powered wheelchairs, folding wheelchairs, standing wheelchairs, sports wheelchairs, etc. Wheelchairs target to promote self-dependence during mobility or walking, increase security and self-confidence and decrease the burden on guardians. Scooters: Scooters are useful for those people with mobility disabilities. Like wheelchairs, a scooter has a seat set on a peak of either four or five wheels. The disabled user puts his feet on foot placements and controls direction using steering wheels. Crutches are medical equipment for those patients who can use their shoulders and arms strength for weight-bearing and not just for stability and balance. Single crutch equalizes 80% burden-bearing support, and both crutches equalize 100% weight-bearing support. Platform, Forearms, and axillaries are three types of crutches. Canes are helpful for people at risk of suddenly falling and having trouble with stability and balance [22]. In contrast, the user walks with a walking stick or cane and feels secure and safe [23]. White cane, forearm cane, and quad canes

84 Advances in Data Science-Driven Technologies

Rani et al.

are common types of canes, and different types, designs and colors of canes are available in the market [24]. Listening and Hearing Aids Hearing-impaired people can hear more clearly by using hearing devices. A hearing aid is a small-size battery-powered tool that wears in a person’s ear or behind it to improve his hearing power [25]. It amplifies sound signals so that he/she can interact and participate in daily activities. It has three components. The sound signals are obtained with the help of a microphone which translates the sound propagation into electrical signals. These signals are passed to an amplifier that amplifies the power of signals and sends them into the ear using a speaker. Behind-the-ear, in-the-ear, and canal hearing aids are styles of hearing aids. ●

●

●

●

Hardwired ALDs (Assistive Listening Devices) use a concrete wire or coil to broadcast the audio signals, binding the listener to the source of the sound. A wired ALD device is typically cheaper, user-friendly, and easy to move. It is the best choice for circumstances where the speaker and listener are not in longer distances, such as in a hotel, an automobile, a theater, or a television viewing. A drawback of the instrument is limited mobility and restrictive seating places determined by the length of the wire. This limits the utility of hardwired devices in larger interaction or communication situations such as in a group discussion, lecture, meeting, or seminar. Wireless ALDs: Wireless devices send the signal without the direct physical connection of a cord between the sound source and the listener. In this technology, a voice transmitter is attached to the source of the sound, and the receiver wears wireless listening devices. The wireless communication between sender and listener can be infrared lights, induction loops, and electromagnetic energy or FM radio waves. Wireless systems can be more movable depending on the power of sound signals and the range of devices and can be restricted by a line of sight and security purpose of the signal. FM system: FM is a wireless ALSs (Assistive Listening System) that uses radiocast technology. It is frequently used in the academic world and offers portability, mobility, adjustability, and flexibility when fit on movable bodyworn transmitters. FM system works with good adaptability and elasticity for all listening circumstances, but it is necessary that the sender and receiver work on an identical frequency [26]. Closed Captions and Subtitles: The individual with a hearing problem can watch TV news, programs, and movies by reading subtitles. By reading subtitles, the users can build their vocabulary and reading skills.

Technology

Advances in Data Science-Driven Technologies 85

Cognitive Devices Cognition is identifying, recognizing, and understanding via thought and processing information. It refers to the cognitive operations of the brain, such as memorizing, planning, creating ideas, reasoning, and problem-solving. Long-term stress, Brain injuries, dementia, social isolation, intellectual impairment, childhood abuse, mental illness, and bereavement are some of the many states that may affect a person's intellectual ability. .

Cognitive devices such as educational software, alarm, calendars, lists, computers, word prediction software, diaries, or electronic assistive devices (e.g., pagers, mobile phones, etc.) help people with cognitive impairment with attention in recognizing, decision-making, problem-solving, matching, reasoning, perceptual skills, association, memorizing etc. Comforting Aids A person with cognitive impairment can feel good or happy by holding his favorite objects such as a toy, a cup, a favorite color bed sheet, a blanket or other personal object. Pets can also assist in making someone feel comfortable or happy. They can cheer, pacify, support, and comfort people. Another device for supporting and comforting people with cognitive impairment is the “Squeeze Machine”. The machine is designed to calm oversensitive or intolerable people, generally those with an autistic spectrum disorder. Limit Motor Skills Aids The people with limited strength in their hands, or those with no hands, use different body parts, such as the forehead, elbow etc., to do their work. ●

●

●

Adaptive Switches: Adaptive switches enable people with physical disabilities to operate and use any switched-enabled equipment like computers, mobile phones, or tablets. The physically disabled person can use the head, toe, thumb, chin, forehead, and elbow to press these switches, whatever fits the person with a physical disability. Mouth Stick: A mouth stick is a stick that is placed in the mouth of a physically disabled person so that she or he can touch the phone screen, play games on the computer, and even press keys on a keyboard. A person with neurological disease or spinal cord injuries can use a mouth stick to perform various activities such as drawing, turning pages, lifting clothes, playing games, etc. Environmental Control Devices: Nowadays, the physically disabled person

86 Advances in Data Science-Driven Technologies

●

●

●

Rani et al.

can easily use Environmental Control Devices if he or she can use some different input mode so that they can easily control electronic instruments at home such as heaters, cookers, fans, tube-lights, and electronically controlled doors. Personal Emergency Response System: A personal or Medical Emergency Response System is a tool that can assist a person with physical disabilities. Usually, this equipment comes in the form of a wearable wristband, pin, or clip attached to clothes, or transmitters worn around the neck or in the pocket. By pressing a button on the equipment, the individual can send an alert signal to the selected caretaker in case of any type of emergency. Speech and Voice Recognition Devices: Speech and voice recognition are valuable devices for differently-abled people who cannot enter data into a computer by typing keys from the keyboard or touching a touch screen. Some special speech and voice recognition software make it possible to insert data and instructions into the computer by only talking to them [27]. Google Home: Amazon Echo or Google Home is an assistant device for those people who can limit the use of their hands, fingers, and arms to be able to use their game devices, computers, and phone. These tools can perform a set of everyday tasks such as making movie recommendations, playing a song, making an invitation, and responding to basic queries expressed by the client about the event.

Vision Aids Visual impairments or low vision significantly impact an individual’s skill to do everyday activities. A variety of assistive technology devices (cheap to expensive) can be used to build confidence and increase participation and self-determination, including the Braille system, mobile phones, radio, low vision lamps, screen readers, white canes, magnifiers, large and bold font print books, Glare Control Devices, Telescopes, translation system, Closed Circuit Television(CCTV), etc. ●

●

●

Optical magnifiers: Optical magnifiers magnify or enlarge the image formed on the eye’s retina. For example, dome, hand-held, hanging, illuminated stands, spectacles, telescopes, pocket magnifiers, etc. Low vision lamps: Enhanced, or good lighting may help people with low eyesight to read and write documents easier, thereby improving their writing and reading performance. These devices include pure vision2, compact fluorescent (CFL) Miroco LED, Tao Tronics LED, halogen lamps and portable desk lamps, etc. Various brightness levels will be helpful in writing, creating art, and reading. The low-vision lamps can be easily found at various malls, online stores, or in catalogs. GPS Locator: GPS locator is a device that is helpful for low vision or visually

Technology

●

●

●

●

●

●

●

●

Advances in Data Science-Driven Technologies 87

impaired people. By using this device, visually impaired people can travel independently without fear of making the wrong turn or getting lost. Braille Translator Software (BTS): Braille conversion software converts electronic data into Braille format and passes it to a Braille embosser which generates a printed copy of the original data. BTS can read a range of digital text files, for example, plain text documents, MS Word, RTF files, PDF, log files, and HTML files [28]. The common conversion software contains Euler, Duxbury Braille Translator, Dolphin Easy Converter, Index Braille app, Braille 2000 US, etc. Screen Readers: Screen magnifiers of enlargement are the applications that help those with low vision read data on the screen [29]. Screen enlargement enlarges the text as we write them to make it more comfortable to read [30]. Audio Format Materials (AFM): It is very helpful for numerous people and students with low vision and visual impairments. This format enables the students to get or read the materials, information, and documents via hearing devices, e.g., I-Pod, CD Players, DAISY (Digital Accessible Information System), computer software, etc. Eye Trackers: Eye-tracking is a device that follows the action and movement of the eyeballs, and eyelids and allows disabled people to use their computer, or mobile by movements of their eyes [31]. The individual who is disabled can type data by looking at key symbols or data presented on the visible screen. But everyone cannot purchase this device because it is very expensive. Talking watches are helpful devices for blind or visually impaired students who are learning day’s names and time scheduling skills. Numerous brands of talking watches in a variety of designs, colors, and sizes are available online, in stores etc. Digital audio recorder: It may assist those students who are visually impaired or blind and can record lectures, e.g., ICD-PX470, Roland R07, Zoom H4n Pro, Olympus WS-852, etc. Talking calculators are specially designed for blind students studying in academic classes. Talking calculator reads aloud each operator, and number key that the student press, and also speaks the result to the question. White cane: It is a mobility device for people with vision loss or blind for safe and independent travel. Blind people can identify uneven surfaces, edges, steps, pits, dents, etc. Some users pick to use a white cane to walk from one place to another, while others use it to tell people about their blindness or visual disability. Various kinds of sticks, like guide cane, mushroom tip, long cane, symbol canes, and all-white sticks, are used for various purposes.

88 Advances in Data Science-Driven Technologies

Rani et al.

Home Security and Safety There are plenty of devices that provide a safe and secure life for disabled people at home. Smart locks, doorbells, smart windows, and small gadgets lift the security system to their homes and reduce the anxiety of their guardians. These devices also help the guardians to monitor the visitors that might be visiting them. Smart lighting, power plugs, carbon monoxide detectors, fire and smoke alarms, alert sentry systems, and water overflow sensors are very useful for helping disabled individuals live independently for longer [32, 33]. Daily Living Aids Self-helping aids help the person with impairments in everyday living activities such as cooking, bathing, washing, dressing, eating, home maintenance, drinking, etc., including pen holders, dressing aids, bath chairs, bathtub seats, eating utensils, adapted clothing, adapted books, powered lift, specialized feeding spoon, grooming, transfer board, folding chair, shower chair, alarm, toileting accessories, time scheduled aids, specialized handles. Various types of daily living aids are shown in Fig. (3) Common subclasses are also explained below. Wheeled shopping trolley Seat belt reacher

Illuminated magnifier

Kev turner

Pill organiser Sock/tights aid One touch can opener

Reacher/grabber

Pedal exerciser

Fig. (3). Daily living aids [34].

●

Cooking and eating aids: AT provides different types of devices for meal preparation, including spread boards, Slicers, scales, cooking baskets, and chopping aids. These devices are specially designed to assist anyone with limited power in their hands or strength to securely and independently cook their own food. But, before selecting cooking and eating aids, you should identify your own personal requirements. Because some devices are more useful for people with one hand and other devices are more suitable for those with limited mobility in the wrist or hands [35].

Technology ●

●

●

Advances in Data Science-Driven Technologies 89

Dining aids: Dining and drinking aids benefit people who face intricacy while eating and drinking. Dining aids include easy-grip utensils, rotating plates, easyhold forks and spoons, a mug with a built-in straw, a steady spoon, a scoop dish plate, a plate guard, etc. These devices help users to eat and drink by themselves. For example, for a person with a weak grip who has difficulty using a spoon, adaptive dining devices such as an easy-hold spoon or fork may be beneficial for him to feed himself independently. Dressing and clothing aids: Dressing and clothing adaptive devices are uniquely designed for people with restricted dexterity, bending precautions, restricted reach, limited mobility, and lack of strength in hands can make it tough to put on the dress [36]. AT offers a variety of dressing aids, including dressing sticks, button hooks, hip replacement packs, zipper ties, ring zippers, elastic shoe laces, etc. Toileting and showering aid: Bathing and toileting equipments for disabled people can make it feasible for them to take comfort showering without any dependence on guardians or caregivers, and privacy in the toilet and bathroom [37]. A wide range of assistive devices, including elevated toilet seats, toilet transfer benches, toilet safety frames, splashguard, bath lifts, grab bars, bathroom trolleys, shower commodes, grab rails, etc., are available in stores and online.

Computer Access Aids People with disabilities can access the computer at home, school, and office, enabled by hardware and software products. It includes adapted or alternate keyboards, touch screens, specialized translators speech to text, light pointers, adaptive mouse, modified keyboards, text readers, head pointers, etc., that enables disabled people to use the computer. Table 1. Different types of assistive devices for disabled people. Disability Devices

Mobility

Vision

Hearing

Communication

Cognition

Manual devices

Walker, wheelchair, corner-chair, walking frame, walking-stick, hand-splint, leg splint, standing frame, walking stick with seat, ferrules shoes,

White cane, magnifier, Braille system, Braille chess, Braille slate, Eyeglasses, low-vision lamp, Braille compass,

Headphones, hearing aid tubes, microphone

Communication cards, boards, communication with signs and pictures.

Adapted keyboards, adaptive mouse, modified keyboards, head pointers, toys, games

90 Advances in Data Science-Driven Technologies

Rani et al.

(Table 1) cont.....

Disability Devices

Mobility

Artificial stair-climbing Intelligence wheelchairs, (AI) based bionic devices prosthetics, limbs

Vision

Hearing

Communication

Cognition

OCR (Optical Character Recognition) system, talking calculator, talking watch, artificial eyes, Voice Over, KNFB Reader

Doorbell, amplified telephone, vibrating alarm clock, hearing loop, amplified cordless phone, visual alert.

Electronic communication equipment with recorder

voice recognition, text readers, touch screens, light pen, trackball, specialized translators speech to text, light pointers,

The Table 1 above represents a wide range of assistive devices for disabled people [38].

MOBILE APPS FOR ALL DISABILITIES A Mobile App is a kind of software application specially devised to execute on hand-held devices such as Android phones, watches or tablets. Mobile apps provide those similar services to users that are accessed on laptops or computers. A wide range of mobile apps has been developed that not only assist normal users but also disabled people in performing their day-to-day activities. Mobile apps help disabled people to enhance their self-confidence, management of accounts, traveling, and social communication [39]. ●

●

●

Voice4u AAC (Augmentative and Alternative Communication): This is a picture-based app specially designed for people facing communication challenges. This app has more than 180 pre-loaded symbols to assist users with speech-challenged. With this app, you can correctly understand the needs and desires of your loved ones and remove the interaction gap. Voice4u app allows the users to extend the built-in library by adding more icons with their own recorded voice. Be My Eyes: This app is particularly designed for blind and low-vision users to assist them in finding their lost things, visiting new places, shopping at malls or stores, etc. Be My Eyes linked the blind user with the sighted volunteer through a video call. The sighted volunteer assists that individual in negotiating unknown environments. This tool is free of cost to use and accessible on both android and IOS. Subtitle Viewer: This app displays open subtitles in real-time on iOS devices. It is beneficial for people with hearing disabilities. They can enjoy TV programs by reading subtitles without the assistance of other family members. Deaf people can learn the spelling of words, understand technical terminology, and increase their reading skills by reading subtitles. Normal people also watch TV shows through subtitles in noise-free environments, like offices, hospitals, libraries, etc.

Technology

●

●

●

●

●

●

●

●

Advances in Data Science-Driven Technologies 91

This tool covers a large number of users because it supports twenty languages. Speak for Yourself (SfY): SfY app was built by speech therapists. This app is being downloaded and used by a large number of users worldwide with limited speaking strength, preschool kids, autism disorder, and cerebral palsy. ModMath: ModMath application was built for parents of kids with intellectual disabilities. This app assists disabled children with dysgraphia, dyscalculia, and dyslexia to easily solve mathematics problems through iPhone. It provides a virtual grid graph that allows the learner to solve math questions on a graphical paper that is comfortably readable for them. But besides basic mathematics calculations, this application also handles large and complicated equations. NotNav GPS Accessibility: NotNav app is specially designed for blind and visual-impaired users. This application is more advanced than traditional GPS tools by declaring your direction, turn, compass heading, address of the nearest street and all details of nearby places. It is simple to use and free of charge. Tap to talk: It is a brilliant mobile that can speak words for someone with speech problems. The user just taps on the words that are present on the display screen, and the tool verbalizes the words on behalf of the user. For people who have speech or communication troubles, this tool could build their confidence to face their daily life activities. Roger Voice app: In the modern era, everybody has a smartphone and enjoys telephone calls with their loved ones. But the deaf people cannot make telephone calls because of their hearing disability. That is why deaf people feel a more difficult life than normal people. To overcome this problem, an app came is known as the Roger voice app. This app converts what an individual says on the telephone call into real-time transcription on the individual's screen. Today, deaf people enjoy telephone calls using the Roger Voice app. Voice Dream Reader: This app is specially designed for students with learning disabilities. This tool provides many reading or learning options, including sentences paragraph, chapter vise, adding special notes, pronunciation of words, etc. The app supports other features such as audio synchronization, a reading speed controller, better color animations, user-friendly text, font size, and many file format options. This app supports more than 20 languages and is available for Android and iOS devices. Learn Braille alphabet: It is an advanced app for blind people. Using this app, the kids, and students can learn how to listen, pronounce the words, and write 26 Braille alphabets. This is specially developed with learner interaction, a userfriendly layout, and good color combinations. The app is gentle and simple to use. Light Detector: Light detectors support blind users in doing their everyday activities independently. It is very simple to run. Just open the app and place your iPhone camera at your desired source or direction. The user will listen to a

92 Advances in Data Science-Driven Technologies

●

●

●

●

●

●

●

●

Rani et al.

high-volume or low-volume sound depending on the intensity of light. The user will listen to a high-volume or low-volume sound based on the strength of the light. Color App: This is developed for blind or low-vision people by Research Center for Technology Development for Differently Abled Persons, Punjabi University Patiala. By using this app, blind people can gain knowledge. This app is simple to use. By focusing the camera on the object, blind people get to know the color of the object. Spoken AAC: The spoken app uses a predictive technique to solve communication problems for people with language and speech disabilities. Not only limited to basic words and phrases, the tool predicts the next terms you are expected to use to create full sentences. You can quickly write full sentences on your iPhone, iPod, and android smartphone. Divyang Sarathi: The main target of this app is to provide all important information related to Persons with Disabilities (PwD), e.g., different schemes, Acts, Policies, Rules, job opportunities, and disability services in an attractive and accessible format. The other features of this application such as audio notes, adjustable font size, user-friendly layout, and bilingual (both English and Hindi). P3 App: This app allows the user to make a video call on his mobile via mobile data or Wifi (Wireless Fidelity) connection. It is designed for the hearing community, targeting to make communication simple. Now deaf people can communicate through the P3 app with their friends, family members, etc. With the P3 app, you not only talk but also see your loved ones who live anywhere. Spread the Sign: This app is the world's biggest sign dictionary because it consists of more than 300,000 signs in 20 different languages. All the signs have been presented by the deaf to increase the accuracy of communication. Spreading the sign is beneficial not only for the deaf community but also for normal people. You can learn many signs in different languages at home using the spread sign application [40]. Learn ISL: This app is developed by Research Center for Technology Development for Differently Abled Persons, Punjabi University Patiala, for deaf and normal people to learn ISL. In this app, the basic words are translated into human sign videos. This app is interactive to learn ISL because it also consists of games. The learner can learn ISL easily and quickly through this game mode. Wheelmate: This app is used to locate wheelchair-accessible restrooms and parking spaces at your nearby locations. Wheelmap: It is an interactive and beneficial app for people with mobility impairments. This app allows mobility-impaired people to locate wheelchairaccessible hotels, cafes, restaurants, theaters, restrooms, malls, and clubs. The Wheelmap app allows users to leave comments and upload photos of accessible places with others. The users can also put marks for more accessible and less

Technology

●

●

●

●

●

●

●

●

Advances in Data Science-Driven Technologies 93

accessible locations for the upcoming times by using the get engaged function. Ghotit Real Writer: It is specially designed for those people who face difficulty in English writing and spelling correction due to dyslexia and dysgraphia disabilities. This app automatically checks and corrects wrongly spelled words, grammatical rules, structure formation, and punctuation. This tool is designed with many features, such as a grammar checker, word prediction, default font size, option for text sharing, and a dedicated dictionary along with explanations. Spy Sam Reading Series app: It is dedicated to students with dyslexia. Learning to read is a difficult task for those students with dyslexia. This app makes learning to read very interestingly by including adventure stories. Every page has restricted but simple vocabulary words. The app mainly interacts with the text, which is why the kids mainly concentrate on reading text rather than pictures. Make sentences: The tool is a small and creative app to build sentences in a wide range of ways; pick words and fit them in the sentences where you see more suitable. This app is an educational and amazing app for children who want to learn sentence structure and improve their English reading and writing skills simply. Book Creator: This app brings creativity among the children to make their poetry books and interactive stories and convey their final result in the form of iBook. The story can be fairy tales, king-queen tales, adventure tales, picture albums, or comic books. This app allows the children to design anything that they think or feel. Book Creative tool is more popular and beneficial for children with autism disorders. Look at me: Kids with autism disorders face challenges in identifying their emotions and making continuous eye contact with other people. Look at me tool targets to assist better communication skills in kids with autism disorder. The kids learn to read the emotions of a person, recognize faces, read an individual’s face, and click photos themselves in different styles and poses through this app. Medication Reminder: This app is extremely helpful and acts as an organizer and reminder for tablets and medication. This tool is really valuable for those people who often forget to take their vitamins and medication at the proper time. By following simple and easy steps, you can put the name of your pills and obtain reminders for taking medicines. Millions of users use Medication Reminder to manage their medicine as prescribed [41]. Finger Print Magic: People with dementia can keep their brains sharp and inventive daily by using the Finger Print Magic app. This app is simple and enjoyable for anyone. You can relax your mind by painting with a combination of lots of colors using your fingers on the phone. Care Zone: The main target of the care Zone app is to create and organize

94 Advances in Data Science-Driven Technologies

●

Rani et al.

health information in a simple and easy way. You can create a medication plan for yourself, your loved ones, your parents, and anyone who suffers from dementia. This app supports many features, such as keeping track of your appointments with a doctor and noting down your new medicine and doctor’s guidelines. Autism Care Skills: This tool supports children with autism disorder in learning interaction, academic, emotional, and social skills. The app provides a wide range of academic learning levels, such as sorting words, color concepts, and simple word spelling. Children with autism disorders learn these academic concepts simply and with more fun by using the autism core skills app.

BENEFITS OF ASSISTIVE TECHNOLOGY DEVICES IN INDIVIDUAL'S LIFE A disabled person can do his work using assistive technology, and he/she does not need to depend on others. AT improves the quality of individual life of the disabled person and reduces the anxiety of guardians [42]. Today, many people with disabilities have good educations, independent life, and participate in home, school, social activities, etc., through technology. They are breaking the barriers in their path and getting more confidence by using technology. Parents with disabilities can take care of their children by using AT devices. A blind mother could use Braille layout to assist her child with a math assignment. A paralyzed mother could walk with her child by using a wheelchair. A father with hard of hearing could use a hearing aid to discuss a reading lesson or story with his kid. Students unable to move their hands can type documents and even draw using a mouth stick. AT contributes to special education, providing equal opportunities and helping disabled students to attain their goals. In short, we can say that AT teaches disabled people how to live in the world. CONCLUSION This chapter presents deep information regarding impaired people suffering from enormous disabilities. Today’s growing technologies have made a great contribution for specially-abled people to make their lives self-dependent and more enjoyable in society. Various devices and mobile apps have been invented and launched in the market for people who need a little bit of help to do their dayby-day activities. People with low vision or visual impairments are helped by Eyeglasses or Braille format, and people with hearing disabilities are assisted by hearing aids, etc. In short, we can say that the assistant devices provide the working scenarios and make them independent.

Technology

Advances in Data Science-Driven Technologies 95

CONSENT OF PUBLICATION The authors consent to publish this book chapter in “Computer Assistive Technologies for Physically and Cognitively Challenged Users” by Bentham Science. CONFLICT OF INTEREST The author declares no conflict of interest, financial or otherwise. ACKNOWLEDGEMENT The author is very thankful to Dr. Vishal Goyal, Professor, Department of Computer Science, Punjabi University, Patiala, and Dr. Lalit Goyal, Associate Professor, DAV College Jalandhar, for their kind guidance and supervision throughout the present work. They helped me and provided ideas to complete the book chapter. REFERENCES [1]

R. Erdem, "Students with special educational needs and assistive technologies: A literature review", Turk. Online J. Educ. Technol., vol. 16, no. 1, pp. 128-146, 2017.

[2]

D. Maor, J. Currie, and R. Drewry, "The effectiveness of assistive technologies for children with special needs: a review of research-based studies", Eur. J. Spec. Needs Educ., vol. 26, no. 3, pp. 283298, 2011. [http://dx.doi.org/10.1080/08856257.2011.593821]

[3]

M.A. Hersh, and M.A. Johnson, "Disability and Assistive Technology Systems", Assist. Technol. Vis. Impair. Blind People, no. May, pp. 1-50, 2008. [http://dx.doi.org/10.1007/978-1-84628-867-8_1]

[4]

J.H. Love, A qualitative study on the challenges faced by entrepreneurs living with physical disabilities within the Sebokeng Township of South Africa, 2016.

[5]

R.M. Gillies, K. Knight, and A.J. Baglioni Jr, "World of Work: perceptions of people who are blind or vision impaired", Int. J. Disabil. Dev. Educ., vol. 45, no. 4, pp. 397-409, 1998. [http://dx.doi.org/10.1080/1034912980450403]

[6]

N.S. Mboshi, Teaching Learners With Visual Impairment in an Inclusive Education Setting: the Cameroon Perspective., 2018.https://www.ijern.com/journal/2018/February-2018/11.pdf

[7]

R.O. Adebisi, N.A. Liman, and P.K. Longpoe, "Using Assistive Technology in Teaching Children with Learning Disabilities in the 21st Century", J. Educ. Pract., vol. 6, no. 24, pp. 14-20, 2015.

[8]

J. Chuaqui, and D. R. Wilson, Other Mental Disorders in Children, 2019.

[9]

A. Sadath, S. Kumar, and S. Mathew, "Mental disorder and disability: A cross-sectional study of disability variance in severe mental disorders", Indian J. Soc. Psychiatry, vol. 34, no. 1, p. 52, 2018. [http://dx.doi.org/10.4103/ijsp.ijsp_2_17]

[10]

S. Halder, A. Talukdar, and M. Pharm, Nature and Causes of Locomotor Disabilities in India. J. Am. Acad. Spec. Educ. Prof, 2013, pp. 48-62.https://eric.ed.gov/?id=EJ1135560 [Online]

[11]

J. Sarkar, D. Dutt, and A. Dasgupta, "Disability among new leprosy patients, an issue of concern: An institution based study in an endemic district for leprosy in the state of West Bengal, India", Indian J. Dermatol. Venereol. Leprol., vol. 78, no. 3, pp. 328-334, 2012.

96 Advances in Data Science-Driven Technologies

Rani et al.

[http://dx.doi.org/10.4103/0378-6323.95449] [PMID: 22565433] [12]

S.K. Bansal, "A study on language disorders in learners, Ayan An Int", Multidiscip. Ref. Res. J., vol. 6, no. 4, pp. 175-185, 2018.

[13]

G.N. Nikisha, and G.A. Menezes, "Hemophilia and its treatment: Brief review", Int. J. Pharm. Sci. Rev. Res., vol. 26, no. 1, pp. 277-283, 2014.

[14]

A. Alias, and N.M. Salleh, "Analysis Of Problems Faced By Special Education Teacher In Teaching The Multiple Disabilities Students", Journal of ICSAR, vol. 1, no. 1, pp. 60-67, 2017. [http://dx.doi.org/10.17977/um005v1i12017p060]

[15]

V. Kumar, "Acid Attacks in India: A Socio-Legal Report", Dignity: A Journal on Sexual Exploitation and Violence, vol. 6, no. 1, 2021. [http://dx.doi.org/10.23860/dignity.2021.06.01.05]

[16]

M. Konecki, S. Lovrencic, and K. Jervis, "Overview of Problems That Students With Disabilities Encounter in Their Higher Education 1", Int. J. Manag. Appl. Sci, no. 2, pp. 2394-7926, 2016.

[17]

S. Narayanan, "A study on challenges faced by disabled people at workplace in Malaysia, Int. J. Stud. Child. Women", Elder. Disabl., vol. 5, no. Oct, pp. 85-92, 2018.

[18]

J.P. Akpan Ph, "D. and L. A. Beard Ed.D., Overview of Assistive Technology Possibilities for Teachers to Enhance Academic Outcomes of All Students", Univers. J. Educ. Res., vol. 1, no. 2, pp. 113-118, 2013. [http://dx.doi.org/10.13189/ujer.2013.010211]

[19]

R.E. Cowan, B.J. Fregly, M.L. Boninger, L. Chan, M.M. Rodgers, and D.J. Reinkensmeyer, "Recent trends in assistive technology for mobility", J. Neuroeng. Rehabil., vol. 9, no. 1, p. 20, 2012. [http://dx.doi.org/10.1186/1743-0003-9-20] [PMID: 22520500]

[20]

S.R. Faruqui, and T. Jaeblon, "Ambulatory assistive devices in orthopaedics: uses and modifications", J. Am. Acad. Orthop. Surg., vol. 18, no. 1, pp. 41-50, 2010. [http://dx.doi.org/10.5435/00124635-201001000-00006] [PMID: 20044491]

[21]

B. Choudhury, and A. Bose, Mobility Aids, 2006. [http://dx.doi.org/10.5005/jp/books/10013_13]

[22]

M. Masum Billah, Z. Mohd Yusof, K. Kadir, and A. Malik Mohd Ali, "Sensory Substitution for Visual Impairments: A Technological Review", What We Know What We Have to Know, no. December, 2020. [http://dx.doi.org/10.5772/intechopen.89147]

[23]

A. Lachtar, A. Kachouri, and T. Val, "Real-time monitoring of elderly using their connected walking stick", In: 2017 Int. Conf. Smart. Monit. Control. Cities, SM2C 2017, 2017, pp. 48-52. [http://dx.doi.org/10.1109/SM2C.2017.8071257]

[24]

P. Arefin, S. Habib, A. Arefin, and S. Arefin, "A comparison of mobility assistive devices for elderly and patients with lower limb injury: Narrative Review", Int. J. Aging Heal. Mov., vol. 2, no. 1, pp. 1317, 2020. [http://dx.doi.org/10.6084/m9.figure share.12318608.v1]

[25]

Frush Holt, "Assistive hearing technology for deaf and hard-of-hearing spoken language learners", Educ. Sci. (Basel), vol. 9, no. 2, p. 153, 2019. [http://dx.doi.org/10.3390/educsci9020153] [PMID: 34113551]

[26]

J.S. Kim, and C.H. Kim, "A review of assistive listening device and digital wireless technology for hearing instruments", Korean J. Audiol., vol. 18, no. 3, pp. 105-111, 2014. [http://dx.doi.org/10.7874/kja.2014.18.3.105] [PMID: 25566400]

[27]

F.R. Sharma, and S.G. Wasson, "Speech recognition and synthesis tool: assistive technology for physically disabled persons", Int. J. Comput. Sci. Telecommun, vol. 3, no. 4, 2012.

[28]

P. T. D. S. H and W. W. K. I. L, "Optical Braille Translator for Sinhala Braille System : Paper

Technology

Advances in Data Science-Driven Technologies 97

Communication Tool Between Vision Impaired and Sighted Persons", Int. J. Multimed. Its Appl., vol. 10, no. 1/2/3, pp. 29-43, 2018. [http://dx.doi.org/10.5121/ijma.2018.10303] [29]

P.R. Singh, Blind Handicapped Vs. Technology : How do Blind People use Computers? 2 S O HOW DOES A BLIND PERSON ACTUALLY KNOW, 2012.

[30]

A. Bhowmick, and S.M. Hazarika, "An insight into assistive technology for the visually impaired and blind people: state-of-the-art and future trends", J. Multimodal User Interfaces, vol. 11, no. 2, pp. 149172, 2017. [http://dx.doi.org/10.1007/s12193-016-0235-6]

[31]

M. M. V. Gatdula, "Assistive technology for locked-in syndrome patients using eye tracker", Int. J. Adv. Trends Comput. Sci. Eng., vol. 9, no. 1.1, pp. 119-124, 2020. [http://dx.doi.org/10.30534/ijatcse/2020/2291.12020]

[32]

N. Chaitra, C. S. Ashwini, and S. Zaidi, "A research on home automation for elderly and physically challenged people by using IOT", Int. J. Innov. Technol. Explor. Eng., vol. 8, no. 6, no. , pp. 473-475, 2019.Assistive technology for locked-in syndrome patients using eye tracker, vol. 8, no. 6, no. , pp. 473-475, 2019. [http://dx.doi.org/10.35940/ijitee.F1098.0486S419]

[33]

S. P. PANDE and P. SEN, "Review On: Home Automation System For Disabled People Using BCI", IOSR J. Comput. Sci., vol. 2014, pp. 76-80, 2014.

[34]

"Trofholz, J. S. Stang, and M. N. Laska, Impact of Cooking and Home Food Preparation Interventions Among Adults: Outcomes and Implications forFuture Programs,", J. Nutr. Educ. Behav., vol. 46, no. 4, pp. 259-276, 2014.

[35]

M. Reicks, A.C. Trofholz, J.S. Stang, and M.N. Laska, "Impact of cooking and home food preparation interventions among adults: outcomes and implications for future programs", J. Nutr. Educ. Behav., vol. 46, no. 4, pp. 259-276, 2014. [http://dx.doi.org/10.1016/j.jneb.2014.02.001] [PMID: 24703245]

[36]

J. Sharawat, and A. Hooda, Special Clothing for Physically Disabled and Elderly, vol. 5, no. 1, pp. 1925, 2018.

[37]

E.L. Friesen, D. Theodoros, and T.G. Russell, "Assistive technology devices for toileting and showering used in spinal cord injury rehabilitation – a comment on terminology", Disabil. Rehabil. Assist. Technol., vol. 11, no. 1, pp. 1-2, 2016. [http://dx.doi.org/10.3109/17483107.2014.984779] [PMID: 25399923]

[38]

A. Technology, and C. Opportunities, Assistive Technology for Children with Disabilities: Creating Opportunities for Education, Inclusion and Participation A discussion paper. World Heal. Organ, 2015, p. 34.

[39]

E.E. Abdallah, and E. Fayyoumi, Assistive Technology for Deaf People Based on Android Platform, 2016. [http://dx.doi.org/10.1016/j.procs.2016.08.044]

[40]

M. Hilzensauer, and K. Krammer, "a Multilingual Dictionary for Sign Languages: ‘Spreadthesign", In: 8th Int. Conf. Educ. Res. Innov., 2015, pp. 7826-7834.

[41]

D. Ameta, K. Mudaliar, and P. Patel, "Medication Reminder and Healthcare – An Android Application", International Journal of Managing Public Sector Information and Communication Technologies, vol. 6, no. 2, pp. 39-48, 2015. [http://dx.doi.org/10.5121/ijmpict.2015.6204]

[42]

A. Ahmed, "Perceptions of using assistive technology for students with disabilities in the classroom", Int. J. Spec. Educ., vol. 33, no. 1, pp. 129-139, 2018.

98

Advances in Data Science-Driven Technologies, 2023, 98-116

CHAPTER 5

Technologies for Hearing Impaired People Using Indian Sign Language Synthetic Animations Rakesh Kumar1,*, Lalit Goyal2 and Vishal Goyal3 Department of Computer Science, University College Miranpur, Patiala, India Department of Computer Science, DAV College, Jalandhar, Punjab, India 3 Department of Computer Science, Punjabi University, Patiala, Punjab, India 1 2

Abstract: This chapter discusses various technologies developed for deaf people using Indian sign language synthetic animations. An automatic translation system for English Text to Indian Sign Language synthetic animations in the real domain has been developed, which consists of a parsing module that parses the input English sentenceto-phrase structure grammar representation on which Indian sign language grammar rules are applied to reorder the words of the English sentence. The elimination module eliminates the unwanted words from the reordered sentence. Lemmatization is applied to convert the words into the root form. The words (or their synonym in case the word is not available in the database) in the sentence are replaced by their HamNoSys code. In case the word or its synonym is not present in the lexicon, the HamNoSys code will be taken for each alphabet of the word. The HamNoSys code is converted into SiGML tags, which are sent to the animation module, which converts the SiGML tags into synthetic Animation using an avatar. Prototypes for announcement systems for deaf people at railway stations, airports and bus stands have been developed. The announcements are categorized and sent to the system in written form. These announcements are dynamically converted to ISL sentences and then animated using HamNoSys and SiGML tags. These translation and announcements systems are the only systems in the country that use continuous synthetic animations of the words in the sentence. Existing systems are limited to the conversion of words and predefined sentences into Indian sign language, whereas our conversion system converts English sentences into Indian sign language in the real domain.

Keywords: Cued Speech, HamNoSys, Hearing impaired people, Indian Sign Language, Lemmatization, Parsing, SiGML, SiGML Stanford Parser, Stemming, Synthetic Animation, Translation System, Visual special Language. Corresponding author Rakesh Kumara: Department of Computer Science, University College Miranpur, Patiala; E-mail:[email protected] *

Manoj Kumar M V, Immanuel Azaad Moonesar R.D., Ananth Rao, Pradeep N, Annappa, Sandeep Kautish and Vijayakumar Varadarajan (Eds.) All rights reserved-© 2023 Bentham Science Publishers

Technologies

Advances in Data Science-Driven Technologies 99

INTRODUCTION There are almost 7117 living languages in the world, which are organized into 153 language families. One of the 153 families of languages used by deaf people to communicate is sign language. This language family includes 144 different sign languages from all around the world, depending on the locality (SIL International 2021). Almost 72 million individuals worldwide are deaf or hearing impaired, out of a total population of nearly 7.5 billion. Only about 4.3 million individuals use Sign language out of such a large number. The remaining almost 67 million deaf and hard of hearing persons do not communicate using correct sign language. As a result, approximately 90% of deaf people have limited or no access to schooling and other informational needs [1]. Deaf people communicate using different hand shapes, facial expressions, finger formations, mouth gestures, and movement of other body parts [1]. The signer uses the 3D space around the body to convey an event, making it a visual-spatial language [2]. Prior to Stokoe's pioneering study, it was thought those sign languages lacked a well-defined structure and syntax. The grammatical rules employed by ISL were discovered by Stokoe's research study, however they are not full-fledged. These signs are only accepted in the small world of deaf people rather than there acceptability in the outside world also. Until the 1960s, sign languages were thought to be nothing more than a collection of gestures and mimes. Dr. Stokoe's research on American Sign Language established that it is a complete language with its own grammar, syntax, and other linguistic characteristics. Other sign languages, such as Indian Sign Languages, are being studied to see if they contain linguistic structure [3]. A sign in Sign language can be made up of both manual and non-manual parts, or both. Hand formations, hand orientation, hand location, and hand movements are the primary components in the manual element of the sign (straight, circular or curved). Facial expressions, eye gazing, body postures and head movements are primary components in the non-manual element of the sign [4]. Some signs, on the other hand, may comprise solely manual or non-manual components. The sign “Yes,” for instance, is made with a vertical head nod and lacks any manual sign. One-handed, two-handed, and non-manual signs are three categories of Sign Language signs. The- Indian- sign- hierarchy- is depicted in Fig. (1).

100 Advances in Data Science-Driven Technologies

Kumar et al.

Fig. (1). Hierarchy of Indian Sign Language Signs.

One-Handed Signs: A single dominant hand is used to depict the one-handed signs. Static or dynamic (having movements) one-handed signals are both possible. Manual and non-manual signs are assigned to each of the static and moving signs. Examples of one-handed static signs with non-manual and manual components are shown in Fig. (2) below.

Fig. (2). One Handed Non-Manual Sign (Headache) and Static Manual Sign (Ear)

Technologies

Advances in Data Science-Driven Technologies 101

Two-Handed Signs: Both of the signer's hands are used to make the two-handed signs. Two-handed signs can be classified in the same way as one-handed signs are classified. Two-handed signs with movements, on the other hand, can be divided into Type 0 and Type1 signs. Both hands are actively involved in Type 0 Signs. One Hand (dominating one) is more actively involved in Type 1 Signs rather than the second hand (non-dominating one), as depicted in Fig. (3) below.

Fig. (3). Two-Handed Sign “Flag” (only the dominating right-hand shows movement) and “long”(both the hands show movement

FACTS ABOUT INDIAN SIGN LANGUAGE Sign language evolves the same as any natural language based on some facts about which people are not familiar or unaware. Some facts about sign languages are as follows: ● ●

●

● ●

●

●

● ● ●

The Sign Languages are NOT similar in all the regions globally. Each Sign Language follows its own grammatical structure rather than just using gestures and pantomime. Sign Languages are confined to a limited vocabulary as compared to other speaking languages all over the world. Unknown words are Finger-spelled in sign languages. Signs of some words may be joined together to form a single sign, e.g., to sign breakfast, the signer must do the signs of Morning and then sign of Food ”Morning+ Food”. The use of adjectives is done after the noun in many of the sign languages, e.g., Coat Blue. Linking Verbs such as “am, are, was, and were” are Never used in Sign Languages. Sign Languages Never make use of “word-endings/suffixes”. Always Present Tense is used to sign the words in Sign Language. Sign Languages don’t make use of articles like “a, an, some, the”.

102 Advances in Data Science-Driven Technologies ●

● ●

●

Kumar et al.

Sign Languages always put WH-questions at the END of the sentence, e.g., “You go where?” Sign Languages don’t make use of gerunds. “(-ing)”. Sign Languages mostly give more weightage to the use of non-manual features/expressions such as the Movement of eyebrows, Movement of eyelids, showing different facial expressions, head movement and movement of the upper torso. Sign Language comes into existence in hearing-impaired and deaf communities.

COMMUNICATION BETWEEN DEAF AND HEARING COMMUNITIES Deaf and hearing impaired communities encounter a lot of problems in common places like post offices, railway stations, airports, hospitals, govt. offices and banks due to a lack of communication. Hearing people are unable to understand any sign shown to them by deaf communities due to a lack of sign interpreters and vice versa. So automatic sign language translation systems that can help deaf and hearing impaired communities understand what is being conveyed at these common places is the need of hour. To bridge the communication gap between hearing and Deaf communities, sign language translation systems must be bidirectional, as shown in Fig. (4).

Fig. (4). Deaf and Hearing Communities’ communication mechanism.

Sign language is a physical movement-based language. It is also called the mother language for deaf people. Sign language is used by a person who is unable to speak or by a person who can hear properly but cannot speak. Sign language is recognized as the first Language for deaf people. In sign language, hands and heads are used to express our thoughts with others. There are a number of sign languages used around the world. Each nation has its own sign language, which is different from others. Observing and understanding sign language is very difficult. It is very difficult for deaf communities to comprehend any kind of information because hearing communities without any sign interpreters cannot get what deaf people want to convey. Therefore, these problems make deaf people isolated from the rest of the world. Deaf people cannot interact with people in common places because their communication is different. Our goal is to help Deaf people in these kinds of situations. Our developed technologies will interpret the text input into three-dimensional avatars.

Technologies

Advances in Data Science-Driven Technologies 103

In sign languages, word-to-word mapping is impossible due to the different word structures between English and sign language. For instance, the word little in English possesses distinct meanings. Little means short, or one can say little knowledge, i.e., not enough knowledge. The same word, “Little” is represented in English, associating with distinct meanings, but the word “Little” is represented, conveying distinct meanings with distinct gestures and signs. Therefore, sign language is not just a representation of the given the word; rather, the distinct meanings are conveyed with the help of Sign language. Finger-spelling is another type representing signs where fingers are used to spell the letters which constitute the word. Finger-spelling is extensively used in parallel with any Sign Language because of the smaller vocabulary of Sign Language. Cued Speech is another alternative to Sign Language. In the cued speech, mouth movements are used along with hand shapes and placement to display/show the sounds. Cued speech works in a similar manner to any speaking language by merging visible hand and mouth gestures to represent sounds just like phonemes [5]. In different parts of India, different types of languages are spoken. Still, many researchers in Indian Sign Language (ISL) have discovered that just one sign language is the primary Language of deaf people. However, different sort of dialects made use of it. Despite variations in SL's lexicons, the grammar rules of all these dialects of SL stay constant. Indian Sign Language is a complete natural sign language with its own syntactical and grammatical structure, phonology, and morphology. ENGLISH TEXT TO INDIAN SIGN LANGUAGE TRANSLATION SYSTEM Knowledge about the lexical and syntactic structure of Indian sign language is necessary for the English text-to-Indian sign language translation system to succeed. The lexicon for “English word to ISL sign” has been developed as illustrated in the following section. The complete system architecture of the sign language translation system is demonstrated in Fig. (5) below. The translation system is built of seven different modules: ● ● ● ●

Text Parser Module to parse the English textual data. Sentence Reorder Module using ISL grammatical rules. Elimination Module to eliminate the unrequired words. Lemmatization Module to find the root form of each word, and the Synonym

104 Advances in Data Science-Driven Technologies

●

●

Kumar et al.

Replacing Module to use synonyms in place of unknown words. Word-to-SiGML Translation Module using Hamburg Notation System Structure. Translation Module using Synthetic Animations.

The system's input is written English text, which is processed to obtain the sentence's phrase structure and grammar representation. The parsed sentence is then submitted to the conversion module, which reorders the English sentence's words according to ISL grammar rules. As English employs the SVO order, and ISL utilizes the SOV order with other variations for interrogative and negative phrases, reordering is required. Unwanted words are deleted from the sentence once it has been formatted according to ISL grammar. This is due to the ISL's usage of just words with some meaning and omission of all aiding words, such as linking verbs, articles, and so on. The lemmatization module receives the output and transforms the words to their root form. This is because, unlike other languages, sign language uses the root form of each word in its phrases, whereas other languages use suffixes, gerunds, and past and future words. Because ISL has a restricted dictionary, unknown words in sentences are replaced with their synonym counterparts; if the synonym is not accessible, the term is finger-spelled character by character. The sentence is now ready to be animated. The English Word-HamNoSys dictionary is used to replace each word in the phrase with its HamNoSys: Writing Notation of the Signs [6] counterpart, and the HamNoSys string is transformed to SiGML (Signing Gesture Markup Language) code using SiGML rules. This SiGML code is transmitted to the SiGML animation tool, which then renders the synthetic Animation.

Fig. (5). Architecture of English Text to ISL Synthetic Animation System

Technologies

Advances in Data Science-Driven Technologies 105

English-ISL Lexicon A bilingual dictionary is required when translating from one Language to another. When an English text is translated into ISL, a bilingual dictionary of English and Indian Sign Language is formed, which includes both the English word and its Indian sign equivalent. The Indian sign corresponding to the English word can be viewed as a real person video, a signed picture, a coded sign language text, or a synthetic animation. All methods have advantages and disadvantages, but synthetic animations are particularly well adapted to translating spoken Language to sign Language. In Table 1 below, you can see a comparison of all the media: Table 1. Comparative analysis of Different Media for Representation of the Signs Media type

Advantages

Disadvantages

Sign Videos

• Realistic • Easy to create

• It takes a long time to generate • It uses a lot of memory • The translation system doesn't support it

Sign Images

• Uses a small amount of memory

• It takes a long time to generate images • Lacks realism when compared to videos • It isn't supported by a translation system

Coded Sign Language Text

• Low Memory Requirement • Supported by a translation system because it is written and can be processed quickly

• Difficult to read and understand • Must be learned

Synthetic Animations

• It uses very little memory • It is easy to recreate • Supported by a translation system • Avatars can be customized

• Avatars are not as realistic as human sign videos.

This system uses synthetic animations to make signs matching to English words because synthetic Animation backed by the translation system is more realistic than photos or coded text. Each English word requires a written form of the sign to be animated. Researchers have worked hard to develop notation systems that allow a three-dimensional sign to be written down. Because synthetic Animation, as opposed to photos and coded text, is far more realistic, synthetic animations have been created for the appropriate English term in this translation system. A textual form of the symbol is used to produce the Animation of each English word. Despite the fact that a 3D sign cannot be written, researchers have worked hard to develop a notation system that allows a three-dimensional sign to be expressed in written form. Stokoe Notation [7], SignWriting [8], Hamburg

106 Advances in Data Science-Driven Technologies

Kumar et al.

Notation System [9], and others are written forms for three-dimensional signs. The dictionary was created using the HamNoSys (Hamburg Notation System) notation [10]. Thomas Hanke developed a HamNoSys alphabet with roughly 200 symbols that cover practically all hand shapes, hand placement, hand/palm orientation, hand movement, and non-manual parts of the sign (Unicode of this notation system is accessible) [11]. The HamNoSys has the following basic structure, as shown in Fig. (6).

Fig. (6). HamNoSys Notation Structure

Every English word is translated to HamNoSys Notation, as shown in Fig. (7), according to the rules of the Hamburg Notation System. Later, this HamNoSys can be transformed into SiGML code, which can then be animated using an Avatar by an animation tool, as seen in the architecture.

Fig. (7). Architecture to produce the Animation from the English words.

The faculty of Disability Management and Special Education [12] has collected the most commonly used words, and a video dictionary has been created. A list of

Technologies

Advances in Data Science-Driven Technologies 107

1818 of the most regularly used English words by differently-abled people was used to create a bilingual dictionary of English words and HamNoSys notation in this research study. Parts of speech are assigned to the words, which are then coded into HamNoSys (FDMSE) Text Parser Module to Parse English Sentences The grammatical structure of the source language is necessary for rule-based language conversion so that the words of the source sentence can be reordered according to the grammar rules of the target language. Parsing is a technique for determining a sentence's grammatical structure. Parsing is used to determine the grammatical structure of English sentences using third-party software. For this research work, we employed the Stanford parser [13], which uses unlexicalized PCFG (probabilistic context-free grammar) and has an output accuracy of 86.36 percent. The probabilistic parser is trained on hand-parsed sentences and then used to parse new sentences using the knowledge obtained. Part-of-speech tagged text, context-free phrase structure grammar representation, and type dependency representation are all possible outputs of the Stanford parser. The Stanford parser parses the English text using Penn Tree tags. We employed phrase structure grammar representation in this project because, in a rule-based approach, the grammatical structure of the English sentence was necessary to transform it into the target language's grammatical structure (Indian sign language). Grammatical Rules for Transformation of English to ISL Sentence When both languages have different grammar rules, translating one spoken Language to another is a difficult task. When the source language is spoken, and the target language is sign language, the complexity is multiplied by a factor of ten. A comparison of the grammar of both languages, given in Table 2, is required while translating an English text into Indian sign language: Table 2. Comparative analysis of ISL Grammar and English Grammar English Grammar

ISL Grammar

Being well-structured, a sufficient amount of research ISL was created for deaf people, and there has been work has been done to define the English grammar very limited research into ISL grammar. ISL rules. It works in the following order: sentences are structured in the following order [14]: “Subject-Verb-Object” “Subject-Object-Verb” Depending on the type of sentences, the English Language employs a variety of verbs and adjectives. In addition, English sentences use a lot of word inflections.

In most of the situations, ISL employs the root form of the word rather than the inflections (gerunds, suffixes, or other forms).

108 Advances in Data Science-Driven Technologies

Kumar et al.

(Table 2) cont.....

English Grammar

ISL Grammar

The English Language has a bigger dictionary than other languages.

There are just about 6000 words in the Indian sign language lexicon (FDMSE, 2021).

In English, the question word comes first in interrogative statements.

The question word is always a sentence ending in Indian sign language [15].

In English sentences, many helping verbs, articles, and conjunctions are utilized.

There are no conjunctions, articles, or linking verbs in Indian sign language.

To transform an English sentence to an ISL sentence, all of the verb patterns [16] are investigated, and rules are created to transform an English sentence into an ISL sentence. The parsed sentence is sent into this module, where the noun and prepositional phrase are frozen, but if the sentence does have a verb phrase, it is checked recursively since the verb phrase might be made up of a noun phrase, verb phrase, prepositional phrase or even the entire sentence. Table 3 below lists some of the conversion rules: Table 3. Examples of Grammatical Reordering of Words of English Sentences Verb Pattern

Rule

Input Sentence

Parsed Sentence

Output Sentence

verb + object

VP NP

come home

(VP (VB come) (NP (NN home)))

Home come

subject + verb

NP V

dogs bark

(NP (NNS dogs)) (VP (VBP bark))

dogs bark

subject + verb + subject complement

NP V NP

(NP (PRP$ her) (NN sister)) her sister (VP (VBD became) (NP (DT a) became a doctor (NN doctor)))

her sister, a doctor, became

subject + verb + indirect object + direct object

NP V NP NP

(NP (FW i)) (VP (VBD gave) I gave him my (NP (PRP him)) (NP (PRP$ my) car (NN car)))

i him my car gave

subject + verb

subject + verb

show him your teeth

(VP (VBP show) (NP (PRP me))) (NP (PRP$ your) (NNS hands))

him your teeth show

NP V NP he cooked food PP for all of us

(NP (PRP he)) (VP (VBD cooked) (NP (NN food)) (PP (IN for) (NP (NP (DT all)) (PP (IN of) (NP (PRP us))))))

The food for all of us cooked

subject + verb + direct object + preposition +prepositional object subject + verb + indirect object + direct object

V NP PP

subject + verb + preposition prepositional object

NP V PP

(VP (VB show) (NP (PRP$ show your teeth your) (NNS teeth)) (PP (TO to) to him (NP (PRP him)))) He is looking for Naresh

your teeth to him show

(NP (PRP he)) (VP (VBP is) He, for Naresh (VP (VBG looking) (PP (IN for) is looking (NP (NN Naresh)))))

Technologies

Advances in Data Science-Driven Technologies 109

Eliminator Module for Removal of Undesired Words Sentences in Indian sign language are made up of major words. There are no linking verbs, articles, or suffixes in the sentence. The ISL phrase is formed once the grammatical rules are applied, and all undesired words must be eliminated. Parts of speech that are not part of an ISL sentence are identified and removed from the sentence. The various parts of speech that do not form part of an ISL sentence, Out of 36 POS tags, are TO, MD(Modals), POS(possessive ending), CC(coordinating conjunction), FW(Foreign word), NNS, NNPS(nouns plural, proper plural), JJR, JJS(adjectives, comparative and superlative), some DT(determiners like a, an, the), RP(particles), Interjections, SYM(symbols), and non-root verbs. The ISL sentence gets free of the above-mentioned undesirable terms. Table 4 below shows examples of unwanted terms that have been eliminated. Table 4. Removal of not required words English Sentence as Input After Phrase Reordering Come home

Final ISL Sentence (post removal of unwanted words)

Home come

Home come

Dogs bark

Dogs bark

Dogs bark

Her sister became a doctor

Her sister a doctor became

Her sister doctor became

I gave him my car

I him my car gave

I him my car gave

Show him your teeth

Him your teeth show

Him your teeth show

He cooked food for all of us He food for all of us cooked

He food all we cooked

Show your teeth to him

Your teeth to him show

Your teeth him show

He is looking for naresh

He for naresh is looking

He naresh looking

Lemmatization and Synonym Replacement The root words are used in Indian sign language sentences. There must be no suffixes, gerunds, or inflexions in any of the words that are utilized. If a word in an ISL sentence is not a root word, it is changed to one after it is passed to the stemmer, and lemmatization rules are applied. Stemming is done with a porter stemmer [17]. The ISL phrase contains only the root words after converting the inflections of the words to their corresponding root words. Each root word is now tested in the English-ISL dictionary for availability. Despite the fact that this dictionary only comprises 1478 words, a list of synonyms is developed to strengthen the system

110 Advances in Data Science-Driven Technologies

Kumar et al.

and increase the number of hits in the dictionary. Rather than compiling a list of all English synonyms, an indirect approach is employed. Only 1478 words have had their synonyms collected (the words which are in our bilingual dictionary). Our database contains roughly 4000 synonym terms. It was considered to eliminate word duplication and the parts of speech of each word. For instance, the word inaugural is an adjective; its equivalent is opening as an adjective, but the word opening can also be employed as a verb. The verbs irate, insense, enrage, and infuriate are all synonyms for anger(v). Anger(n) also has synonyms such as annoyance, irritation, anger, rage, resentment, and hostility, all of which are nouns. The word is spelled if it cannot be found in a dictionary or synonym list. In this case, spelling the word entails finger-spelling, which entails taking each character of the word and determining which sign will be created. The database now has signs for all of the English alphabet, bringing the total number of words in the dictionary to almost 6000. All personal nouns (names of people, places, and things) are finger-spelled. Sign Animation using Avatar It becomes ready to be animated after all the processing on the English sentence has been completed to convert it to an ISL sentence. Synthetic Animation (using a computer-produced avatar known as an Avatar) is most suited for creating the sign, as previously stated. We used the Animation Programme SiGML Player to animate the sentence [18]. The input for this tool is SiGML (Signing Gesture Markup Language) tags, and the output is animations of the Avatar performing the signs.

Fig. (8). Methodology to generate the synthetic animations from the English word.

Each word of the ISL sentence is substituted with its HamNoSys code to generate the SiGML tags. Each character (alphabet) of the word is replaced with the matching HamNoSys code of the corresponding letter for words that are not in the English-ISL dictionary database. The HamNoSys code for the entire ISL phrase is

Technologies

Advances in Data Science-Driven Technologies 111

now available. Using HamNoSys-SiGML conversion rules, the HamNoSys code is now translated to SiGML tags. When we have all of the SiGML tags for the entire sentence, we send it to the animation tool, which turns it into an animated Avatar. The complete Methodology to generate the synthetic animations from English words is depicted in Fig. (8). ANNOUNCEMENTS SYSTEM FOR RAILWAY STATIONS In terms of size, Indian Railways is the world's fourth-largest railway network. In the year 2018-19, a total of 8,439 million passengers were served the services of travel on trains at 7321 railway stations [19]. Apart from these many numbers, the metro railways and other local area trains work independently. Navigating railway stations is very difficult for members of the Deaf community because no station in India uses ISL for announcements. A prototype has been developed for announcements at railway stations [20], in which the announcements have been divided into three categories, as shown in Fig. (9).

Railway Announcements

Static

Dynamic

Special

Fig. (9). Categorization of Railway announcements.

Static announcements are those which are fixed and do not change. These announcements are free from factors like time, train numbers, train names, platform numbers, etc. We have collected 67 such announcements. In our developed protocol, these announcements are preloaded. Playing a particular announcement is just selected, and the Avatar plays it. Dynamic announcements include the time (in hours and minutes), train number, train name, platform number, source station name, destination station name, via station name, etc., within it. We have collected the dynamic announcements and divided them into ten categories: Arriving, Arriving shortly, Available, Departure, Departure shortly, Expected to arrive, Running late, Special train, Empty train, and Train cancel. In all these categories, the changing factors are time, train

112 Advances in Data Science-Driven Technologies

Kumar et al.

number, train name, platform number, source station name, destination station name, and via station name. In our developed protocol, these announcements are generated dynamically by selecting the dynamic fields: time (in hours and minutes), train number, train name, platform number, source station name, destination station name, via station name, etc. Special announcements can also be called instant announcements. These announcements are not available in the database; rather, they are generated instantly on special occasions, like during accidents or any natural calamity. In reality, these announcements are spoken instantly by human beings as per the situation. In our proposed protocol, these announcements can be produced in ISL using the Text to ISL conversion system, which has already been discussed in detail in the previous section. ANNOUNCEMENTS SYSTEM FOR AIRPORTS India is the world's third-largest domestic civil aviation market after the United States and China. The Airport Authority of India (AAI), a government-owned company, operates 449 airports, including 126 commercial airports. Approximately 341 million passengers are handled by these commercial airports [21]. Announcements at the airports of India do not use ISL, so deaf people face many problems at the airports. A prototype has been developed for announcements at airports [22]. The announcements of the airports may be static or dynamic. Again, these announcements are categorized as shown in Table 5: Table 5. Examples of Airport Announcements Sr. No

English Sentence

1

Pre-Flight Announcements Kindly switch off all personal electronic devices, including laptops and mobile phones.

2

Dynamic

Gate Change Announcement You are requested to proceed to gate number_____ for boarding

5

Dynamic

Inflight Announcement Our flight time will be of ____hours and ____ minutes.

4.

Static

Pre-Boarding/Final Boarding Announcements This is the last boarding call for passengers ______ (name) and ____ (name) booked on flight _____(No.) to _____(City).

3.

Announcements Type

Refreshment Announcement

Dynamic

Technologies

Advances in Data Science-Driven Technologies 113

(Table 5) cont.....

Sr. No

6

English Sentence

Announcements Type

Passengers travelling to_____ (destination) on _______ flight ______ (flight number) are requested to proceed to_____ (area) for refreshment.

Dynamic

Paging Announcement May I have the attention of Mr/Ms.________ travelling on flight _____ to _______

7

Delay Announcement We are sorry for this delay and sincerely apologize for any inconvenience caused to you.

8

Static

Post Flight Announcement The temperature outside is _____degree Celsius

11

Static

Safety Briefing Announcement To operate your seatbelt, insert the metal tip into the buckle to tighten and pull the loose end of the strap

10

Static

Security Announcement Passengers are requested to kindly pass through Security Check and await departure announcement in the ground-floor/first-floor security area

9

Dynamic

Dynamic

Miscellaneous Announcements Have a pleasant flight

Static

The corresponding Indian Sign Language translation of the above announcements was done with the help of an Indian Sign Language Interpreter. The ISL translation of the above announcements was being used in the developed system and shown to various deaf school teachers and students. The response from their side was very satisfactory as they were able to understand announcements being displayed using synthetic animations. ANNOUNCEMENTS SYSTEM FOR BUS STANDS India is a big country having 29 states covering a major area of Asia. Each state runs many bus stands. A system prototype of the Bus Stations System for hearing–Impaired to announce bus stations announcements or instructions having a rule-based MT approach is proposed. It is the first-ever MT system being proposed in the public domain of bus stations for easy and understandable conversion of all the announcements/instructions being made at bus stations into Indian Sign Language (ISL) synthetic animations. This proposed MT system prototype accepts inputting of all the instructions/announcements in the English text and converts them equivalent Indian Sign Language (ISL) synthetic animations as output using the

114 Advances in Data Science-Driven Technologies

Kumar et al.

corpus of a developed bilingual English-ISL made for the public domain. Announcements at bus stations in India do not use Indian Sign Language. Therefore, hearing-impaired people face many problems at bus stations. A system prototype has been developed for announcements at bus stands/terminals. The announcements of the bus stands may be static or dynamic. Some sample announcements are given in the following Table 6: Table 6. Examples of Bus Stands Announcements Sr. No

English Sentence

1

Bus Arrival Announcements Passengers, your attention, please. Bus number........... From.............. via………… to.................... is just arriving shortly on Counter number ……….

2

Dynamic

Security Announcements Please do not leave your luggage unattended.

4.

Dynamic

Bus Departure Announcements Passengers, your attention please. Bus number..............from ............to……… will leave shortly from counter number …………..

3.

Announcements Type

Static

Instructional Announcements Do not throw waste material around; please use the dustbins provided.

Static

CONCLUSION AND FUTURE WORK This chapter presents various technologies developed for better communication for Deaf/hearing-impaired people. The automatic translation system converts the English text to Indian sign language synthetic animations and is the first ever translation system for Indian sign language in the real domain. Currently, the system has been created for context-free conversion of simple English sentences to synthetic animations. Human evaluation is the best method for MT evaluation, and if the MT system is from text to Sign Language, it becomes almost essential to test the system on human beings. This is because the grammar of Indian Sign Language is still not standard, and the Language used varies to some extent depending on the region. Due to lack of time, cost, and, most importantly, the interpreters, it becomes impractical. Overall conversion accuracy has been checked by demonstrating the system in various deaf schools. After evaluating our system on 647 simple sentences, it is found that the overall accuracy of English text to the Indian Sign Language Machine Translation system is 82% on the basis of the accuracy test.

Technologies

Advances in Data Science-Driven Technologies 115

The announcement systems using ISL synthetic animations for Railway stations and Airports have a strong potential to benefit the Deaf community. As per the available information, the announcement system for railway stations and airports will be the world's first announcement system. We are committed to working on some other projects which might prove useful for deaf people. We are currently working on ●

●

●

●

Development of an automated system for news telecast by animated Avatar through Indian sign language for deaf people. Development of a system for automatic translation of complex and compound English sentences to Indian sign language synthetic animations. Machine translation system from Punjabi text to Indian sign language synthetic animations Automatic conversion of human sign video to SiGML tags using machine learning technique.

CONSENT FOR PUBLICATION Not applicable. CONFLICT OF INTEREST The Figures used in this chapter are in the public domain (copyright-free) and are under other open licenses. Hence, no consent is to be taken from anyone for the materials used in this chapter. ACKNOWLEDGEMENT This chapter and the research behind it would not have been possible without JASigning software, which has been made downloadable and can be used for evaluation and individual research purposes. JASigning has copyright UEA(20052014), and its development was supported by the EU Framework 7 project DictaSign. We are very much thankful to SAAR Municipal Services Management Company for supporting us financially in the development of an automatic conversion tool for public announcements at airports in Indian sign language synthetic animations. We are very much thankful to LET’SIGN for their linguist support of sign language. We acknowledge the teachers and students of the deaf schools of the region who have supported us in the guidance and evaluation process of our systems.

116 Advances in Data Science-Driven Technologies

Kumar et al.

REFERENCES [1]

Accredited Language Services, Cued Speech., 2016.Cued Speech., 2016.

[2]

https://www.accreditedlanguage.com/2016/08/17/setting-cued-speech-apart-from-sign-language/

[3]

D.M. Eberhard, F.S. Gary, and D.F. Charles, Ethnologue: Languages of the World. Twenty-fourth edition. Dallas, Texas: SIL International., 2021.

[4]

https://www.ethnologue.com/guides/how-many-languageshttps://www.ethnologue.com/subgroups /sign-language

[5]

Faculty of Disability Management and Special Education (FDMSE).http://indiansignlanguage.org /dictionary

[6]

JASigning, http://vh.cmp.uea.ac.uk/index.php/JASigning

[7]

Ministry of Railways, CMS Team., 2021. http://www.indianrailways.gov.in/railwayboard/

[8]

M.F. Porter, Snowball: A language for stemming algorithms., 2001. http://snowball.tartarus.org /texts/introduction.html

[9]

Timir Mozumder, A language for stemming algorithms., 2021. http://www.knowindia.net /aviation3.html

[10]

World Federation of the Deaf, https://wfdeaf.org/

[11]

L. Goyal, and V. Goyal, "Automatic Translation of English Text to Indian Sign Language Synthetic Animations", 13th International Conference on Natural Language Processing., pp. 144-153, 2016.

[12]

T. Hanke, "HamNoSys - Representing sign language data in language resources and language processing contexts", LREC 2004, Workshop Proceedings: Representation and Processing of Sign Languages., 2004 pp. 1-6. http://www.sign-lang.uni-hamburg.de/dgs-korpus/files/inhalt_pdf/ HankeLRECSLP2004_05.pdf

[13]

D. Klein, and C.D. Manning, "Accurate unlexicalized parsing", Proceedings of the 41st Annual Meeting on Association for Computational Linguistics-Volume 1, Association for Computational Linguistics., pp. 423-430, 2003.

[14]

R. Kumar, V. Goyal, and L. Goyal, Airport Announcement System for Deaf., 2020. https://aclanthology.org/2020.icon-demos.15 a

[15]

R. Kumar, V. Goyal, and L. Goyal, Railway Stations Announcement System for Deaf., 2020. https://aclanthology.org/2020.icon-demos.16 b

[16]

S. Sinha, A skeletal grammar of Indian sign language., 2003.

[17]

W.C. Stokoe Jr, "Sign language structure: an outline of the visual communication systems of the American deaf. 1960", J. Deaf Stud. Deaf Educ., vol. 10, no. 1, pp. 3-37, 2005. [http://dx.doi.org/10.1093/deafed/eni001] [PMID: 15585746]

[18]

W.C. Stokoe Jr, and M. Marschark, "Sign language structure: an outline of the visual communication systems of the American deaf. 1960", J. Deaf Stud. Deaf Educ., vol. 10, no. 1, pp. 3-37, 2005. [http://dx.doi.org/10.1093/deafed/eni001] [PMID: 15585746]

[19]

V.J. Sutton, Lessons in SignWriting Textbook., 2014.Http://Www.Signwriting.Org/

[20]

P.C. Wren, and H. Martin, High school English grammar and composition. S. Chand & Company Pvt. Ltd., 1999.

[21]

U. Zeshan, "Indo-Pakistani Sign Language grammar: a typological outline", Sign Lang. Stud., vol. 3, no. 2, pp. 157-212, 2003. [http://dx.doi.org/10.1353/sls.2003.0005]

[22]

U. Zeshan, M. Vasishta, and M. Sethna, "Implementation of Indian Sign Language in educational settings", Asia Pac. Disabil. Rehabil. J., vol. 16, no. 1, pp. 16-40, 2005.

Advances in Data Science-Driven Technologies, 2023, 117-134

117

CHAPTER 6

Augmentative and Alternative Communication/ Hearing Impairments Jestin Joy1,*, Kannan Balakrishnan2 and M Sreeraj3 Department of Computer Applications, St. George’s College, Aruvithura, Kerala, India Department of Computer Applications, CUSAT, Kerala, India 3 Sree Ayyappa College, Eramallikkara, Alappuzha, Kerala, India 1 2

Abstract: Data-driven technologies aid in effective communication for deaf people. Research on sign language recognition, sign language generation and tools based on them is going at a fast pace. With the easy availability of depth sensors, specialized data sets, efficient machine learning algorithms, and computational capabilities provided by specialized hardware, the development of efficient data science-based solutions for deaf users and people with difficulty in hearing is possible now. This chapter focuses on recent research on Automatic Sign Language Recognition (ASLR), Sign Language Production (SLP) and tools based on them. A major focus of this chapter is research and tools using Sign Languages since they are the most commonly used communication medium by deaf people. Research on sign languages from different parts of the world as well as the effectiveness of Machine Learning techniques for ASLR and SLP, are discussed in detail.

Keywords: Augmentative and Alternative Communication (AAC), Automatic Sign Language Recognition (ASLR), Avatar, Corpus, Deaf, Depth Sensors, Generative Adversarial Network (GAN), Gesture, Indian Sign Language (ISL), Kinect, Leap Motion, Machine Learning, Neural Network, Pose, Recurrent Neural Network(RNN), Recognition, Sign Language, Sign Language generation, Sign Language Production (SLP), Vision, Variational AutoEncoder (VAE). INTRODUCTION Augmentative and Alternative Communication (AAC) refers to the communication methods used to aid or replace speech or writing for those with spoken or written language impairments. According to estimates, around 2 million people use AAC. AAC can be aided or unaided. In an unaided approach, no * Corresponding author Jestin Joy: , Department of Computer Applications, St. George’s College, Aruvithura, Kerala, India; E-mail:[email protected]

Manoj Kumar M V, Immanuel Azaad Moonesar R.D., Ananth Rao, Pradeep N, Annappa, Sandeep Kautish and Vijayakumar Varadarajan (Eds.) All rights reserved-© 2023 Bentham Science Publishers

118 Advances in Data Science-Driven Technologies

Joy et al.

external tools are used. Sign Language is an example of unaided communication. Aided approaches make use of some devices for communication. This can range from simple ones like pen and paper to complex speech-generating devices. Sign language is one of the major communication mediums for deaf people. It involves gestures for communication. There are different sign languages around the world. Not only deaf people but parents of deaf people, researchers, social workers etc., also need to learn sign language. Since it consists of gestures, sign language is not easily represented in printed media. Two things are important for communication: translating gestures to text and vice versa. These two things are difficult because gestures are difficult to process using current techniques. Sign language gestures are made up of manual and non-manual components, making communication difficult. Though tools exist that help in a limited domain, full-fledged communication systems are an active area of research. Accuracy, ease of use and affordability are some important considerations in selecting sign language-based assistive tools. Understanding the adoption of assistive technology tools is critical for its design and development. Studies have shown that adoption rates of assistive technology tools are very low. A major problem facing hardware-based assistive products developed is the high cost incurred. This bars people from using it. Even if these assistive tools reach their intended audience, most of them are abandoned over a short period of time. Studies show that the abandonment rate ranges from 8% to 75%. It has also been observed that, due to the lack of feedback from the users, most people are unhappy with the assistive tools they are using. Data-driven technologies that aid effective communication for deaf people play an important role in developing assistive technologies for the Deaf. This includes techniques for sign language recognition, sign language generation and lip reading identification. With the availability of enough data, efficient applications based on these are possible. Each gesture in Sign Language comprises a number of building blocks, including articulation points (from joints such as the finger, arm, wrist, and elbow), hand configuration, action type, hand direction, and facial expression. Recognising sign language, which consists of finger spelling and varies across languages, is critical. Most sign languages use two-handed gestures, though some of the letters of the alphabet need one hand. Since it requires both hands, feature occlusion and computer vision techniques may fail to extract these features. Most data-enabled research for deaf people revolves around Automatic Sign Language Recognition (ASLR). Various depth-enabled sensors are used to efficiently recognize different sign language gestures. Due to the complexities involved in sign language grammar, current systems can only efficiently recognize isolated words.

Communication

Advances in Data Science-Driven Technologies 119

Sign Language Production (SLP) is an emerging research area mainly based on using avatars. With the emergence of Machine Learning based techniques, the generation of natural-looking sign videos is possible. Research has shown that avatars that closely match human gestures are also possible using machine learning techniques. Text and video are the two input modalities for SLP. The linguistic complexity of sign languages makes it difficult to transform text to sign language gestures. Processing RGB videos is also a challenge. The huge amount of data involved makes it computationally intensive. This chapter is organized as follows. After discussing the background of sign languages, sign language recognition, generation techniques and data, science-based AAC solutions are discussed in detail. BACKGROUND Signs in sign language are analogous to morphemes, and the articulations of the hands and body can be described as phonemes. Chereme (in Greek, meaning hand) is used to represent the smallest meaning-bearing word. A sign is a sequential or parallel construction of these smallest meaning-bearing units. Unlike spoken languages, sign languages allow simultaneous signs. The lack of iconicity exhibited by sign language makes learning difficult for early-stage learners. The availability of a corpus plays a major role in the development of any language. Very few corpus projects exist for Sign Languages. RWTH-Phoeni-2014T [1] is one of the most popular datasets used in sign language translation. It consists of data from German public TV weather forecasts. Along with the corpus information, results suitable for translation and recognition tasks like hand tracking, sign recognition and sign language translation are provided by the dataset. Content4all [2] is a much larger dataset from an open domain. It consists of 190 hours of news footage annotated with sign language information. Content4all is a collection of six datasets. Swedish Sign Language Corpus Project, DGS-Corpus project, Corpus NGT project, BSL Corpus Project and How2Sign are some of the other major corpus projects around the world. Like corpus projects, dictionaries also play a major role in language development. Some of the corpus projects, like the DGS-Corpus project, have dictionaries also included with them. HandSpeak, ASL-LEX, LSE-Sign and ISLRTC ISL Signs are some of the sign language dictionary projects from around the world. Different projects exist around the world to help deaf people. Most of these involve providing accessible facilities for deaf people. ViSiCAST(Virtual Signing: Capture, Animation, Storage, and Transmission) is a European Unionfunded project focusing on its citizens. The virtual avatars Simon, Tessa and Visia, which form the basis of earlier projects, also form the foundation for

120 Advances in Data Science-Driven Technologies

Joy et al.

ViSiCAST. eSIGN is another European Union-funded project to provide information in sign language. Avatars are used to provide information. It aims to provide solutions so that developers can integrate them into web pages. A sign language notation system was also developed as part of these projects. DICTASIGN project provides interaction with users through the sign-wiki solution. It was also provided with web pages in mind. Greek Sign Language (GSL), British Sign Language (BSL), German Sign Language (DDGS) and French Sign Language (LSF) are supported by DICTA-SIGN. A major highlight of DICTASIGN is that it provides a sign recognition module making use of a Kinect sensor. SignSpeak [3] project was also proposed with sign recognition in mind. It aims at a continuous translation of sign language to spoken language text. With the availability of recognition datasets, projects focussing on automatic sign language recognition have become available. The contributions provided by these projects to the deaf and research community are immense. A sign language gloss is a typed approximation of it. Sign Language gloss enables the transcription of Sign Language. Though not much practical use, these are helpful to represent sign language as the written text. Sign Language transcription looks at the representation of signed utterances in a two-dimensional, linear format. As forms can be executed simultaneously, signal language transcription is a very challenging task. This is true for recognition also. Stokoe notation, SignWriting and SigML are solutions that aim to solve this problem. These systems are very helpful as they are used as intermediate forms in the signgeneration process. This is discussed in the later section on sign language generation. This is very important since the grammar of a spoken language is different from sign language; intermediate forms help to easily translate spoken language text to many different sign languages. Sign language recognition and generation are two important techniques essential for the efficient communication of deaf people. Machine Learning techniques have introduced efficient mechanisms for both. The following section discusses sign language recognition, generation techniques and data science-based AAC solutions. Sign Language Recognition Scholarly approaches for overcoming disability-related challenges are many, and they differ according to the setting. Sign language recognition (SLR) systems, which are used to convert SL signals into text or voice in order to communicate with those who do not know these signals, are one of the major treatments. SLR systems may be roughly grouped into two: Sensor-based and vision based. This is shown in Fig. (1).

Communication

Advances in Data Science-Driven Technologies 121

Fig. (1). SLR system category.

These methods are used to track hand settings and identify movement. SLR glove systems are one of the most important initiatives for gathering data on human hand movement. A sensor-based system needs sensors, while a vision or imagebased system only requires a camera which is less expensive than the previous type. Both necessitate keeping track of hand motions. Sensor-based System Sign detection depends on extracting data with respect to some parameters. In terms of hardware, the glove-based recognition system consists of taking input, processing and output. This is shown in Fig. (2). The most prominent motion that the four fingers (pinkie, ring, middle, and index) can do is bending towards the palm and then returning to the previous position. The thumb has a notable advantage over the other fingers in that it may freely move in six directions. When it comes to fingers, the most common action in SL is bending. As illustrated in Fig. (2), the gesture data is collected from a specific type of gloves equipped with several sensors, including flexion (or bend) sensors, accelerometers (ACCs), proximity sensors, and abduction sensors through input devices made up of microcontrollers. For sign recognition, features such as finger bend angles, abduction between fingers, wrist orientation (roll, pitch, and yaw), and degrees of freedom (DoF), are extracted using feature extraction algorithms. Depending on the number of sensors in the glove, these gloves can give anywhere from 5 to 22 degrees of freedom.

Fig. (2). Architecture of the glove-based system.

122 Advances in Data Science-Driven Technologies

Joy et al.

Glove-based systems have a substantial advantage over vision-based systems in that gloves may convey critical and needed data (degree of bend, pitch, etc.) to the computing device in the form of voltage values [4], eliminating the need to translate raw data into usable values. Vision-based systems, on the other hand, must apply specific tracking and feature extraction algorithms to raw video streams, which increases the computational load [5, 6]. The wearability aspect is a major disadvantage of glove based system. Vision-based Systems Cameras are the primary tools employed by vision-based systems to collect critical input data in the form of images. This is shown in Fig. (3). The major advantage of utilising a camera is that it eliminates the need for sensors lowering the system's construction costs. Most laptops have high-resolution cameras for capturing data. After obtaining data, a data augmentation process is carried out. Despite the high-resolution camera available in most smartphones [7], there are several difficulties, such as the capturing device's limited field of view, high computational expenses [8, 9], and the need for many cameras to obtain trustworthy results (due to depth and occlusion issues [10, 11]). These issues are inherent in the technology, rendering it difficult to develop real-time recognition applications. To obtain the features, various feature extraction techniques such as Orientation Histogram, Combined Orientation Histogram and Statistical (COHST), Wavelet Features, and local binary patterns (LBP) with and without

Fig. (3). Vision-based system.

conditionality reduction techniques such as principal component analysis (PCA) may be used. Features from sign images are obtained by the algorithm itself in deep learningbased methods for recognition. By using different models, training and validation

Communication

Advances in Data Science-Driven Technologies 123

can be done. Shallow machine learning employs a variety of methods for training and recognition, including Neural Networks (NN), Support Vector Machines (SVM), k-NN, Pulse-Coupled Neural Networks (PCNN), Hidden Markov Models (HMM) and others. In addition, multiple transformation techniques (such as Fourier, Hartley, and Log-Gabor transforms) were used [11] to extract and summarise attributes from a collection of sign frames. Hybrid methodology integrating glove and camera-based technologies is also employed by researchers. This approach involves mutual error elimination to increase overall accuracy and precision. However, because of the high cost and computational overheads of the entire system, little study in this field has been done. Nonetheless, augmented reality systems show promise when paired with hybrid tracking approaches [12]. Challenges and motivation of Sign Language Recognition When humans communicate with one another, they utilise their voices and gestures. Gestures, particularly in the case of those who are deaf or hard of hearing, can either supplement or completely change their communication. However, because so many people use diverse gestures to represent the same notion, people have difficulty interacting with others and regularly find themselves in awkward situations. Fig. (4) illustrates the Challenges and motivation of Sign Language Recognition.

Fig. (4). Challenges and motivation of Sign Language Recognition

124 Advances in Data Science-Driven Technologies

Joy et al.

Commonly used Sensors The flex sensor is the most popular of sensors [13 - 17], which detects the amount of finger curvature and is utilised by a wide spectrum of researchers and developers. Resistive carbon components are used in flex sensor technology. When the substrate is bent, the sensor outputs a resistance proportional to the radius of the bend—the smaller the radius, the greater the resistance. As a result, as the component's body bends, the flex sensor's resistance increases. The flex sensor is incredibly thin and light, making it quite convenient to wear, and it is available in two sizes: 2.2 inches and 4.5 inches. Surface Electromyography (sEMG) sensors measure the electrical potential produced by muscles. When placed on the forearm over big muscles, only particular hand and finger motions may be recognised. Each sEMG sensor has a threshold value that is generally specified, and the muscle is meant to activate when the electrical potential surpasses this threshold. Optics technology is utilised to determine the shape of the finger curvature by measuring the amount of light streaming through the channel [18 - 20]. Optical sensors are electrical detectors that convert light or a change in light into an electronic signal. Furthermore, a combined pair of light-emitting diode-light dependent resistors (LED-LDR) is used to detect finger bends. LDR is a component that has a (variable) resistance that varies with the quantity of light it receives [21]. Both of these optical devices work by determining light density when the finger is straight. Both of these optical devices measure light intensity, thus, while the finger is straight, the density of incoming light is quite high; when the finger bends, the converse is true. A sensitive sensor is a durable polymer-thick film device that alters resistance by applying a force. This sensor measures the strength between 1 and 100 kN. The tactile sensor resistance changes as extra pressure or force are applied. If no pressure is applied, the sensor looks like an open circuit, and the resistance decreases when the pressure is increased [4]. The tactile sensor, therefore, detects if the finger is curved or even by determining the force applied to it [16, 22]. A circular area diameter of a tactile sensor spans between 0.16 and 0.5 to 1 inch. Another technique, using the Magnetic Sensor Hall Effect (HEMS), is to determine the shape of the finger using an applied magnetic field and monitor the voltage fluctuation over an electrical conductor. The unipolar Hall Effect Sensor (MH183) is used to detect and readily access the magnet's southern poles. Effect sensors are put on the fingertips, and the palm is positioned on the magnet with the South pole pointing up.

Communication

Advances in Data Science-Driven Technologies 125

Although SL postures include movement of the hand and wrist, they should be investigated. In addition to detecting the shape of the finger, the ACC sensor's characteristics distinguish the movement and rotation of the pencil as additional consideration. The 3-axis ACC, which gives acceleration differences on each axis, is therefore used to detect the direction and velocity of the handle, allowing the sensor glove to work properly. The ADXL335 (Adafruit Industries, New York, NY, USA) is a small, thin, low-power three-axis accelerometer with voltage signal conditions. It measures velocity and has a total minimum range of 3g. In tilt-sensor applications, this device measures static gravity acceleration as well as dynamic acceleration caused by movement, shock, or vibration. When coupled with a 3-axis ACC and a 3-axis Gyroscope on the same board, it is useful for monitoring tracked movement; the device gives acceleration data in all three directions as well as rotation around each axis [19, 23, 24]. Gyroscopes measure the angular speed and the rate at which an axis spins. An accelerometer may not provide enough information to properly determine the orientation of a moving object. Companies such as Microsoft, Leap Motion and Intel, have developed several ways of detecting and tracking the user's movement by introducing the Kinect and the Leap Motion Controller (LMC), respectively. Kinect can track human movement via an RGB camera, a depth sensor, and a multi-array microphone. Thus it recognises the human skeleton and tracks its hands, while LMC just tracks hands utilising integrated cameras and infrared sensors [25, 26]. The LMC is a low-cost portable peripheral palm device designed for tracking hand and finger movement in 3D Cartesian coordinates, similar to the sensor module. This device is made up of two monochrome cameras and three LEDs [25] and has a field of interaction that is about 8 cubic feet above the equipment. The LMC includes an API that allows users to collect hand-and-finger data such as fingertips, handpainted position, and so on [25]. Different Recognition Models Recognition may be divided into three types depending on machine learning algorithms: unsupervised, supervised, and deep learning. Unsupervised learning includes techniques such as dimensionality reduction, clustering, and ensemble methods. For dimensionality reduction, several methods, such as Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) were used. The clustering technique included k-means, Gaussian Clustering, and SelfOrganizing Feature Maps (SOFM). Various combinations of models, such as neural networks (NN) and hidden Markov models (HMM), decision trees (DT) and NN, DT and HMM, SOFM and NN, and so on, were also used. For classification, algorithms such as Nave Bayes, Decision Trees, Random Forests,

126 Advances in Data Science-Driven Technologies

Joy et al.

Shallow Neural Networks, Support Vector Machines (SVM), and k-Nearest Neighbour (k-NN) were commonly employed. Researchers chose Convolutional Neural Networks (CNN), DeepNeural Networks, Long short-term memory(LSTM), and Recurrent Neural Networks for deep learning. Although decision models based on value, policy, and fuzzy rules were analysed in the context of reinforcement learning. The technique proposed by Razieh Rastgoo et al. [28] proposed a vision-based model of sign language recognition utilising deep learning algorithms, which showed substantial improvement in accuracy. Because CTC (connectionist temporal classification)-based sign language recognition techniques struggle to anticipate output for longer videos, Liqing Gao et al. [29] proposed an RNNTransducer-based SLR architecture. A visual hierarchy transcription network is built in this framework to collect the required details. A lexical prediction network is used to extract effective contextual information from output predictions. Sara Askari Khomami et al. [30] presented a low-cost sign language recognition device. It was built with surface electromyography (sEMG) and Inertial Measurement Unit (IMU) sensors and is utilised by 10 volunteers. When the 25 highest-ranked features of the two modalities (sEMG, IMU) were retrieved and categorised using the kNN classifier, the average accuracy was 96.13%. According to Nayan M.Kakoty et al. [31], identification is accomplished by the use of an indigenously created data glove. The glove is used to measure the angles of the finger and wrist joints, after which the data is filtered and scaled. It is made up of an accelerometer, a microprocessor, and an FSK transceiver module that transmits data between the glove and the sign language recognition unit. Using a label-matching approach, the detected sign language is translated into related speech. Using this approach, average recognition rates of 96.7% were obtained. M.A.Ahmed and colleagues [32] developed a real-time system using wearable sensing gloves. It includes 17 sensors and 65 channels for capturing hand motion data. A 3D-printed humanoid arm was used to validate the sensor. As a consequence, the accuracy for numbers, alphabet letters, and words was 99 percent, 96 percent, and 93.4 percent, respectively. Hasmath Farhana Thariq Ahmed et al. [33] proposed a recognition system based on WiFi. This system can identify 30 static and 19 dynamic sign motions. The suggested system made use of machine learning classifiers such as SVM, kNN, RF, NB, and CNN, as well as a deep learning classifier. A dynamic Persian sign language recognition system was described by Saeideh GhanbariAzar et al. [34]. A simple region-growing approach was used to extract hand-shape information from a collection of 1200 videos. These time-varying trajectories were then modelled with the Hidden Markov Model (HMM) and Gaussian probability density functions as observations. These trajectories were then modelled with a Hidden Markov Model (HMM) and Gaussian probability density functions as observations.

Communication

Advances in Data Science-Driven Technologies 127

Sign Language Generation Compared to sign language recognition, research in sign language generation has not progressed much. Sign language generation is challenging since visual and linguistic characteristics are essential. Initially, manual animation of 3D avatars is used to provide information. This is a very difficult task, considering the complexity of sign language and the difficulty in animation. Annotation of nonmanual features is also a challenge. The lack of non-manual features is cited as a drawback for many existing 3D avatar-based models. Utilizing motion capture data for animation provided better results but was limited in the set of signs that could be provided. Due to this, data-driven techniques provided a promising future. Most works [35 - 37] in sign language generation use RGB videos as input. Each video represents a single sign or a sentence. Extracting individual signs in a sentence is also a challenging task. Processing RGB videos is a computationally intensive task. Extracting skeleton data [37] from RGB videos and using it for sign language generation is also explored by researchers. Convolutional Neural Networks (CNN) can easily extract features from RGB videos which can later be used for sign language production. Features provided by CNN can also be used as input for Recurrent Neural Network (RNN)/Generative Adversarial Networks (GAN) based networks. An overview of works that use these is discussed in this section. When the input is text, sign language production is more complex. Spoken language text to sign language is a very difficult task because of a number of reasons. Since sign languages have visual modality, in sign language conversion, textual data should be converted to a completely different form as far as textual data is concerned. Research on sign language grammar is very little, making the conversion process more difficult. Though sign language possesses all the properties of a spoken language, the complexity of sign language grammar makes conversion difficult. Neural Machine Translation (NMT) techniques can be used for machine translation. S Stoll et al. [35] proposed a Neural Network (NN) based approach for sign language generation. The system proposed by S Stoll et al. comprises three stages. A text-to-gloss network, a gloss-to-pose network and a Variational AutoEncoder (VAE)/Generative Adversarial Networks (GAN) based sign generation network. The authors also introduced SignSynth [36], another data-driven sign language video generation system making use of Generative Adversarial Networks (GAN). NN-based techniques have been shown to have superior results as compared to traditional methods. The architecture of the method is given in Fig. (5).

128 Advances in Data Science-Driven Technologies

Joy et al.

Fig. (5). Architecture of NN-based SL generation proposed by S Stoll et al. [1].

Three networks form the core of this. The first network is text to glose, which converts textual input to sign language gloss. Glose to Pose network takes this input and generates pose. The final network is Pose to Video, which takes pose and outputs sign language video. Pose to video network uses a Generative Adversarial Network (GAN) for sign language video generation. In that, a generator G is engaged in generating video based on the pose and discriminator D tries to differentiate real or fake outputs. As time progresses, G is able to create more realistic sign videos making it difficult for D to correctly classify real or fake outputs. Heike Brock, Felix Law et al. [37] proposed a pipeline for generating skeleton data from Sign Language videos. Single-camera sign language videos are used for the input, and the proposed system could generate three-dimensional skeleton data. These can be used for Sign Language synthesis using a virtual avatar. Generation process is divided into two stages. This is shown in Fig. (6).

Fig. (6). Sign generation process as described in Heike Brock, Felix Law et. al. [37]

Communication

Advances in Data Science-Driven Technologies 129

In the first stage, full-body joint positions are generated using pose estimation techniques. Three Recurrent Neural Networks (RNN) are made used for this. In the second stage, these are fed to the Inverse Kinematics (IK) calculator to generate angular and translational displacements of joints. These can be used for avatar animation. This work uses two types of data are used, 2D joint positions obtained using pose estimation and 3D absolute joint positions obtained from motion capture data. Three RNN’s are used for inference of 3D joint position of face, finger and rest of the points in the global coordinate system. For training of RNN’s, motion capture data corresponding to the sign videos is used as the target. Input is the pose estimated sign video. The study has found that RNN’s could correctly predict the depth information from horizontal and vertical information. First RNN outputs 3D finger points, the second RNN outputs 3D face points, and the third RNN outputs the rest of the 3D points. Upper half of the body is enough for Sign Language generation tasks since not much difference exists for the lower half of the body. B. Saunders, N. C. Camgoz, and R. Bowden [38] proposed a transformer-based model for sign language production. It converts spoken language sentences to 3D sign sequences. It uses two transformer architectures; a symbolic transformer for textual input to gloss and a progressive transformer to translate gloss to 3D poses. They also experimented with a model using transformers to directly translate text to 3D sign sequences. This uses a single progressive transformer model. In their study text to pose network has better accuracy as compared to text to gloss to pose network. Data Science based AAC Solutions Advancements in sign recognition accuracy have opened options for the development of applications for deaf people using machine learning techniques. Though processing facial expressions and full-body movements is still a challenge, applications with a focus on learning are possible [39 - 42]. These solutions are possible for different sign languages and focus on early-stage learning. Most of the solutions make use of Automatic Sign Language Recognition techniques. Authors couldn't find any large-scale use of the sign language recognition-based tools. SmartSignPlay [42, 45] is an application for learning American Sign Language for deaf/hard of hearing and parents of deaf children. A gesture-based learning mechanism using CNN is proposed in it. Evaluation of the prototype developed shows that SmartSignPlay is favoured by participants. Helene Brashear [43] proposed an improvement to copycat games using ASLR for American Sign

130 Advances in Data Science-Driven Technologies

Joy et al.

Language (ASL). The main aim of the project is to help children practise ASL. Using the proposed Sign Language Recognition technique, users can interact with the game using gestures. It proposed a user-independent recognition mechanism for ASL hand gestures. Sensor-based solutions are also tried by researchers. Kinect-sign [44, 46] is a kinect-based solution for sign language learning. Kinect-Sign uses matching techniques for Portuguese Sign Language recognition. Kinect was selected because of its ability to record depth data so that recognition accuracy could be improved. In the Kinect-sign skeleton, data is captured using Kinect and Region of Interest (ROI) is calculated. This is then converted to grayscale, and nearest neighbour interpolation is done for recognition. This mechanism could work when the number of gestures to be recognised is small and fails when the number of gestures increases. Researchers employed various machine learning algorithms since it provides better accuracy on depth data from these types of sensors [47]. CONCLUSION AND FUTURE DIRECTIONS The availability of depth sensors, specialized data sets and efficient machine learning algorithms and computational capabilities provided by specialized hardware made it possible for the development of efficient data science-based solutions for deaf users and people with difficulty in hearing. Since sign language is made up of manual and non-manual components, the development of solutions that solves real word problems is still a challenging task. Results from sign language recognition tasks show that reasonable accuracy can be obtained if the correct sensors and algorithms are used. Compared to this, sign language production is in a nascent stage. Advances in 3D avatars and depth estimation made it possible to model manual and non-manual components in Sign Language. CONSENT FOR PUBLICATION Not applicable. CONFLICT OF INTEREST The author declares no conflict of interest, financial or otherwise. ACKNOWLEDGEMENT Declared none.

Communication

Advances in Data Science-Driven Technologies 131

REFERENCES [1]

J. Forster, C. Schmidt, T. Hoyoux, O. Koller, U. Zelle, and J. H. Piater, "Rwth-phoenix-weather: A large vocabulary sign language recogni-tion and translation corpus", inLREC, vol. 9, pp. 3785-7389, 2012.

[2]

N. Cihan Camgoz, B. Saunders, G. Rochette, M. Giovanelli, G. Inches, R. Nachtrab-Ribback, and R. Bowden, Content4all open research sign lan-guage translation datasets, 2021.

[3]

P. Dreuw, H. Ney, G. Martinez, O. A. Crasborn, J. Piater, and J. Miguel Moya, The signspeak projectbridging the gap between signersand speakers., 2010.

[4]

M. Aarthi, and P. Vijayalakshmi, "Sign language to speech conversion", 2016 International Conference on Recent Trends in Information Technology (ICRTIT), pp. 1-6.

[5]

L.T. Phi, H.D. Nguyen, T.Q. Bui, and T.T. Vu, "A glove-based gesture recognition system for vietnamese sign language", 2015 15th International Conference on Control, Automation and Systems (ICCAS)., pp. 1555-1559, 2015. [http://dx.doi.org/10.1109/ICCAS.2015.7364604]

[6]

T. Chouhan, A. Panse, A.K. Voona, and S. Sameer, "Smart glove with gesture recognition ability for the hearing and speech impaired", IEEE Global Humanitarian Technology Conference - South Asia Satellite (GHTC-SAS)., pp. 105-110, 2014. [http://dx.doi.org/10.1109/GHTC-SAS.2014.6967567]

[7]

R. Suharjito, R. Anderson, F. Wiryana, M.C. Ariesta, and G.P. Kusuma, "Anderson, F. Wiryana, M. C. Ariesta, and G. P. Kusuma, “Sign language recognition application systems for deaf-mute people: A review based on input-process-output,”", Procedia Comput. Sci., vol. 116, pp. 441-448, 2017. [http://dx.doi.org/10.1016/j.procs.2017.10.028]

[8]

A. Alvi, M. Azhar, M. Usman, S. Mumtaz, S. Rafiq, R. Rehman, and I. Ahmed, "Pakistan sign language recognition using statistical template matching, World Academy of Science, Engineering and Technology, International Journal of Computer, Electrical, Automation", Control and Information Engineering, vol. 1, pp. 765-768, 2007.

[9]

P. Kumar, H. Gauba, P. Pratim Roy, and D. Prosad Dogra, "A multimodal framework for sensor based sign language recognition", Neurocomputing, vol. 259, no. 02, pp. 21-38, 2017. [http://dx.doi.org/10.1016/j.neucom.2016.08.132]

[10]

A. Erol, G. Bebis, M. Nicolescu, R. Boyle, and X. Twombly, "Vision-based hand pose estimation: A review", Computer Vision and Image Understand-ing, vol. 108, pp. 52-73, 2007. [http://dx.doi.org/10.1016/j.cviu.2006.10.012]

[11]

A.I. Sidig, H. Luqman, and S.A. Mahmoud, "Transform-based Arabic sign language recognition", Procedia Comput. Sci., vol. 117, pp. 2-9, 2017. [http://dx.doi.org/10.1016/j.procs.2017.10.087]

[12]

V.S. Bhatnagar, R. Magon, R. Srivastava, and M. Thakur, "A cost effective sign language to voice emulation system", Eighth International Conference on Contemporary Computing (IC3)., pp. 521-525, 2015. [http://dx.doi.org/10.1109/IC3.2015.7346737]

[13]

S. Paul, B. Neogi, M. Chowdhury, R. De, and P. Das, Analytical study and overview on glove based indian sign language interpretation technique, vol. 54, no. 6, .

[14]

N. Praveen, N. Karanth, and M. Megha, "Sign language interpreter using a smart glove", 2014 International Conference on Advances in Electronics Computers and Communications, pp. 1-5, 2014. [http://dx.doi.org/10.1109/ICAECC.2014.7002401]

[15]

S.F. Ahmed, S.M.B. Ali, and S. Qureshi, "Electronic speaking glove for speechless patients, a tongue to a dumb", In: 2010 IEEE Conference on Sustainable Utilization and Development in Engineering and Technology, 2010, pp. 56-60. [http://dx.doi.org/10.1109/STUDENT.2010.5687009]

132 Advances in Data Science-Driven Technologies

Joy et al.

[16]

Y-F. Fu, and C. Ho, "Development of a programmable digital glove", In: 2007 International Conference on Machine Learning and Cybernetics vol. vol. 4. , 2007, pp. 1948-1953. [http://dx.doi.org/10.1109/ICMLC.2007.4370466]

[17]

N. Tubaiz, T. Shanableh, and K. Assaleh, "Glove-based continuous arabic sign language recognition in user-dependent mode", IEEE Trans. Hum. Mach. Syst., vol. 45, no. 4, pp. 526-533, 2015. [http://dx.doi.org/10.1109/THMS.2015.2406692]

[18]

N.H. Adnan, K. Wan, A.B. Shahriman, S.K. Zaaba, S. Basah, Z.M. Razlan, D. Hazry, M.N. Ayob, M.N. Rudzuan, and A.A. Aziz, "S. nisha Basah, Z. M. Razlan, D. Hazry, M. N. Ayob, M. Rudzuan, and A. A. Aziz, Measurement of the flexible bending force of the index and middle fingers for virtual interaction", Procedia Eng., vol. 41, pp. 388-394, 2012. [http://dx.doi.org/10.1016/j.proeng.2012.07.189]

[19]

L-J. Kau, W-L. Su, P-J. Yu, and S-J. Wei, "A real-time portable sign language translation system", IEEE 58th International Midwest Symposium on Circuits and Systems (MWSCAS)., pp. 1-4, 2015. [http://dx.doi.org/10.1109/MWSCAS.2015.7282137]

[20]

G. Pradhan, B. Prabhakaran, and C. Li, "Hand-gesture computing for the hearing and speech impaired", Multimedia, IEEE, vol. 15, pp. 20-27, 2008. [http://dx.doi.org/10.1109/MMUL.2008.28]

[21]

C. Preetham, G. Ramakrishnan, S.K. Gonugondla, A. Tamse, and N. Krishnapura, Hand talkimplementation of a gesture recognizing glove Texas Instruments India Educators’ Conference , 2013, pp. 328-331. [http://dx.doi.org/10.1109/TIIEC.2013.65]

[22]

D. Abdulla, S. Abdulla, R. Manaf, and A. Jarndal, "Design and implementation of a sign-t-speech/text system for deaf and dumb people", 5thInternational Conference on Electronic Devices, Systems and Applications (ICEDSA)., pp. 1-4, 2016."Design and implementation of a sign-t-speech/text system for deaf and dumb people", [http://dx.doi.org/10.1109/ICEDSA.2016.7818467]

[23]

"An assistive interpreter tool using glove-based hand gesture recognition", IEEE Canada International Humanitarian Technology Conference - (IHTC)., pp. 1-5, 2014.

[24]

M.I. Sadek, M.N. Mikhael, and H. Mansour, "A new approach for designing a smart glove for arabic sign language recognition system based on the statistical analysis of the sign language", 34th National Radio Science Conference (NRSC)., pp. 380-388, 2017. [http://dx.doi.org/10.1109/NRSC.2017.7893499]

[25]

L. Motion, https://www.leapmotion.com/

[26]

Kinect for Windows, https://developer.microsoft.com/en-us/windows/kinect

[27]

R. Gupta, and A. Kumar, "Indian sign language recognition using wearable sensors and multi-label classification", Comput. Electr. Eng., vol. 90, 2021.106898 [http://dx.doi.org/10.1016/j.compeleceng.2020.106898]

[28]

R. Rastgoo, K. Kiani, and S. Escalera, "Sign language recognition: A deep survey", Expert Syst. Appl., vol. 164, 2021.113794 [http://dx.doi.org/10.1016/j.eswa.2020.113794]

[29]

L. Gao, H. Li, Z. Liu, Z. Liu, L. Wan, and W. Feng, "Rnn-transducer based chinese sign language recognition", Neurocomputing, vol. 434, pp. 45-54, 2021. [http://dx.doi.org/10.1016/j.neucom.2020.12.006]

[30]

S.A. Khomami, and S. Shamekhi, "Persian sign language recognition using IMU and surface EMG sensors", Measurement, vol. 168, 2021.108471 [http://dx.doi.org/10.1016/j.measurement.2020.108471]

[31]

N.M. Kakoty, and M.D. Sharma, "Recognition of sign language alphabets and numbers based on hand

Communication

Advances in Data Science-Driven Technologies 133

kinematics using a data glove", Procedia Comput. Sci., vol. 133, pp. 55-62, 2018. [http://dx.doi.org/10.1016/j.procs.2018.07.008] [32]

M.A. Ahmed, B.B. Zaidan, A.A. Zaidan, M.M. Salih, Z.T. Al-qaysi, and A.H. Alamoodi, "Based on wearable sensory device in 3D-printed humanoid: A new real-time sign language recognition system", Measurement, vol. 168, 2021.108431 [http://dx.doi.org/10.1016/j.measurement.2020.108431]

[33]

H.F.T. Ahmed, H. Ahmad, K. Narasingamurthi, H. Harkat, and S.K. Phang, "Df-wislr: Device-free wi-fi-based sign language recognition", Pervasive Mobile Comput., vol. 69, 2020.101289 [http://dx.doi.org/10.1016/j.pmcj.2020.101289]

[34]

S.G. Azar, and H. Seyedarabi, "Trajectory-based recognition of dynamic Persian sign language using hidden Markov model", Comput. Speech Lang., vol. 61, 2020.101053 [http://dx.doi.org/10.1016/j.csl.2019.101053]

[35]

S. Stoll, "N. C. Camg ̈oz, S. Hadfield, and R. Bowden, “Sign language production using neural machine translation and generative adversarial networks", In: Proceedings of the 29th British Machine Vision Conference (BMVC 2018) British Machine Vision Association, 2018.

[36]

S. Stoll, S. Hadfield, and R. Bowden, "Signsynth: Data-driven sign language video generation", In: European Conference on Computer Vision Springer, 2020, pp. 353-370.

[37]

H. Brock, F. Law, K. Nakadai, and Y. Nagashima, "Learning three-dimensional skeleton data from sign language video", ACM Transactions onIntelligent Systems and Technology (TIST)., vol. 11, no. 3, pp. 1-24, 2020. [http://dx.doi.org/10.1145/3377552]

[38]

B. Saunders, N.C. Camgoz, and R. Bowden, "Progressive transformers forend-to-end sign language production", In: European Conference on ComputerVision. Springer, 2020, pp. 687-705.

[39]

J. Joy, K. Balakrishnan, and M. Sreeraj, "SignQuiz: A Quiz Based Tool for Learning Fingerspelled Signs in Indian Sign Language Using ASLR", IEEE Access, vol. 7, pp. 28363-28371, 2019. [http://dx.doi.org/10.1109/ACCESS.2019.2901863]

[40]

J. Joy, K. Balakrishnan, and S. M, "SiLearn: an intelligent sign vocabulary learning tool", J. Enabling Technol., vol. ahead-of-print, no. ahead-of-print, pp. 173-187, 2019. [http://dx.doi.org/10.1108/JET-03-2019-0014]

[41]

J. Joy, K. Balakrishnan, and S. Madhavankutty, Developing abilingual mobile dictionary for indian sign language and gather-ing users experience with signdict, vol. 32, no. 3, pp. 153-160, 2020. [http://dx.doi.org/10.1080/10400435.2018.1508093]

[42]

C-H. Chuan, and C.A. Guardino, "Designing smartsignplay: An interactive and intelligent american sign language app for children who are deaf or hard of hearing and their families", In: Companion Publication of the 21stInternational Conference on Intelligent User Interfaces, ser. IUI ’16 Companion.New York, NY, USA: Association for Computing Machinery., 2016, pp. 45-48. [http://dx.doi.org/10.1145/2876456.2879483]

[43]

H. Brashear, "Improving the efficacy of automated sign language practice tools SIGACCESS Access. Comput., no. 89, p. 11-17", Online (Bergh.), 2007. [http://dx.doi.org/10.1145/1328567.1328570]

[44]

J. Gameiro, T. Cardoso, and Y. Rybarczyk, “Kinect-sign: teaching sign language to “listeners” through a game,” in International Summer Work-shop on Multimodal Interfaces. Springer, 2013, pp. 141-159.

[45]

S. Rani, and S. Kautish, "Association clustering and time series based data mining in continuous data for diabetes prediction", In: Second international conference on intelligent computing and control systems (ICICCS). , 2018, pp. 1209-1214. pp. 1209-1214. IEEE, 2018. [http://dx.doi.org/10.1109/ICCONS.2018.8662909]

[46]

T. Rudra, and S. Kautish, "Impact of Covid-19 Infodemic on the Global Picture", EAI/Springer

134 Advances in Data Science-Driven Technologies

Joy et al.

Innovations in Communication and Computing, vol. COVID-19, pp. 333-353, 2021. [http://dx.doi.org/10.1007/978-3-030-68936-0_16] [47]

O. Obulesu, "Suresh Kallam, Gaurav Dhiman, Rizwan Patan, Ramana Kadiyala, Yaswanth Raparthi, and Sandeep Kautish. Adaptive Diagnosis of Lung Cancer by Deep Learning Classification Using Wilcoxon Gain and Generator", J. Healthc. Eng., p. 2021, 2021.

Advances in Data Science-Driven Technologies, 2023, 135-163

135

CHAPTER 7

Hardware and Software-based Accessibility Innovations to Help Physically Disabled User Bhagvan Kommadi1,* 1

Director of Product Engineering, Value Momentum, Hyderabad, India Abstract: Initially, text-based terminals were used as computers with DOS OS. The terminal had a cursor, and text was typed into the terminal. Accessibility for users was provided by using the text and events related to the user. The accessibility formats were voice and text in various font levels. The web has evolved, and accessibility support for the web is very important. HTML5 helps us to design websites for PDAs, Mobile, TV browsers, and other devices. Web browsers use the accessibility API for disabled users. Accessibility APIs used in the browsers are MSAA, UIA, and Microsoft UIA. This chapter in the book talks about accessibility software and hardware used in software design and development.

Keywords: Accessibility, Design, Hardware, Platform, Software. INTRODUCTION This book will highlight the accessibility of software and hardware implemented in software projects. Desktop, web, and mobile are designed using various tools to implement accessibility requirements. We will look at different elements of the accessibility design, development, and deployment platforms. Accessibility is important for many users with disabilities, such as blindness, deaf, and others. They have problems like website content, color, images, and contrast. They use screen readers to change the content to voice for the disabled. Popular screen readers are listed below: Jaws, Window-Eye, NVDA, Seortek System Access, Apple VoiceOver, PRCA, BRLTTY, Emacspeak, WebAnywhere Spoken Web ChromeVox, ChromeVis. Screen readers help the users for changing the content to a format that the user can understand and act upon it. Screen readers have features such as text-to-speech, braille display method, and piezo effect-based crystals. You can use other Corresponding author Bhagvan Kommadi: Director of Product Engineering, Value Momentum, Hyderabad, India; E-mail:[email protected]

*

Manoj Kumar M V, Immanuel Azaad Moonesar R.D., Ananth Rao, Pradeep N, Annappa, Sandeep Kautish and Vijayakumar Varadarajan (Eds.) All rights reserved-© 2023 Bentham Science Publishers

136 Advances in Data Science-Driven Technologies

Bhagvan Kommadi

software and hardware assistive techniques for accessibility support Table 1. There are devices like sip and puff switch and others for the disabled user for content typing and reading. Table 1. Accessibility Matrix . Context/Disability

Situation

Vision Mobility

Object holding in the hand

Hearing

Noise

Cognitive

Temporary

Permanent

Head concussion

Blind

arm injury

Repetitive Stain injury

Head concussion

ACCESSIBILITY FOR DIFFERENT DISABILITIES All operating systems have accessibility support, and users can handle content types like images, videos, and audio. OffScreen model for the content is created using API calls, such as retrieving text, images, events, formats, and actions. Screen readers, voice dictation software, and speaking word processors use this offscreen model. The offscreen model can have objects, context, user actions, and OS native calls. This model can be OS-independent, and it can have contentspecific features, such as white spacing, content alignment, and formatting specifications. Accessibility can be not only due to disability but also low bandwidth, low-speed internet, and mobile capabilities. Blind users and users with low-level vision and color blindness might be targeted for accessibility support. Some users might not have limbs or disabilities, which prevent them from typing properly. They can be provided hardware that has voice support for speech-to-text conversion. Audio content can be an issue for deaf users. Captions and voice/video alternatives are provided to deaf users. Accessibility support is provided for other groups of users who might have schizophrenia, dyslexia, depression, and ADHD. Compliance standards have evolved in the accessibility area, such as WCAG, WCAG 2.0 WCAG 2.1, Section 508, and CVAA. ADA act stipulates the content design that needs to target disabled readers. WCAG compliance is used to certify the content for levels like A, AA, and AAA. Accessibility software and tools have features for users who have issues in the following areas: Reading, Writing, Scanning Content.

Accessibility

Advances in Data Science-Driven Technologies 137

Accessibility software needs to support industry standards and best practices [1]. It needs features to handle multiple operating systems, device types, and content types. This software needs to be audited and certified using checklists and best practices. Reports can be generated for accessibility compliance, violations, compliance rates, the severity of the issues, noticeability issues, tractability issues, and recommendations related to accessibility. Accessibility software needs to support the following: content types, controls, input indicators, signals, and alerts. It needs to provide features related to cognitive accessibility, which are: memory issues, processing speed issues, organization, problems coordination problems Accessibility certification and verification is another project in the software lifecycle. It consists of the following phases: Planning, Analysis, Design [2] Development, Testing [3] and Maintenance. This project involves the following areas: website design, structure outline, web page template development, template integration, content creation, and website publishing. It would be best if you had a team for accessibility project execution. Accessibility architects, developers, and testers are needed for design and accessibility-certified website publishing. Testing tools [4] and software are required for verification and certification purposes. Different companies, like Microsoft, sell accessibility-specific tools to verify accessibility requirements. Project managers are needed on the team to manage the complex project related to accessibility. The design of the accessible website will depend on the following key elements: charts, graphs, trees, outlines, page tabs, dialogs, calendar controls, animations, dynamic content, mobile, multi-media, control playback, typography, menus, navigation, page structure, color, contrast, forms, images, keyboard, links, language, typography, frames, data tables, dialogs, authoring tools and layout tables. Accessibility requirements related to text readability are listed below: sentence length, paragraph length, language complexity, content headings, content headers, content footers, content type, UI controls and usage, voice-over features, text hierarchy, bulleted lists, text font types, spelling issues, idiomatic English grammar-related issues. Addressing the text readability requirements helps in avoiding tiring and mental taxation for the users. Session management requirements are as follows: session time outs, expiry, communication channel support, action time for completion. Content-type related requirements are related to the following: user’s interaction avoid confusion, distraction prevention, flashing effects, parallax effects, motion effects, screen size, Page-level settings and global-level settings.

138 Advances in Data Science-Driven Technologies

Bhagvan Kommadi

The issues related to content types such as animations, images, videos, and others are as follows: Seizure, Dizziness, Vertigo motion, and Distractions. These content types need to have captions, subtitles, and transcripts to help disabled users. Icons should not have too much information. Internationalization and localization might create different issues related to accessibility. Cultural differences might result in choosing different colors and themes, which might be an important design factor for accessibility. Navigation and inks need to have the right positions and targets. Automatic refresh is not a good design for accessibility. CSS design needs to factor in the colors and themes suitable for accessibility. Accessibility software-specific content needs to be designed and rendered offscreen. The order of web page elements and the content needs to be mentioned in the correct place. Image maps and images need to have the right Alt Text. Keyboard accessibility design needs to factor in the focus order of the elements. Assistive hardware and software for input entry need to know the element order. Aggregator platforms that gather content and collate it based on categories must factor in the accessibility requirements. The published content must abide by the accessibility guidelines and requirements. Content needs to be scrollable and resizable based on the accessibility design requirements. In the next sections, we are going to look at critical elements in the accessibility ecosystem, designing for accessibility, best practices, digital accessibility, accessibility project lifecycle, planning for accessibility, and accessibility platform. CRITICAL ELEMENTS - ACCESSIBILITY ECOSYSTEM The reader will be aware of the essential factors of the accessibility ecosystem. We will focus on key elements (Fig. 1) which affect accessibility, such as charts, graphs, timber and outlines, web page tabs, dialogs, calendar controls, animations, dynamic content, mobile, multi-media, and typography. Every user and reader has equal rights to the content material and the facts provided. When the user accesses the content material for a unique task, the content material needs to serve the accessibility requirements of a disabled user. The incapacity can be associated with hearing, imagination, and the prescient moves of the body. Tools should be used for making the content material created accessible. Microsoft and different software programs furnish accessibility and precise equipment to repair the documents.

Accessibility

Advances in Data Science-Driven Technologies 139

Fig. (1). Accessibility Ecosystem.

The software program which publishes the content consists of accessibility tags Table 2. Authors should exhibit the impact on the conversation channel. The channel can be a picture, audio, or video. Accessibility and best assurance are very important steps in the publishing process. Table 2. Accessibility Tags Access/Device Monitor

Colorblind palette

Screen reader

Printer

Braille embosser

Haptic Device

Plotter

Keyboard

Yes

No

Yes

Yes

Yes

Yes

Yes

Mouse

Yes

Yes

Yes

Yes

Yes

Yes

Yes

JoyStick

Yes

No

No

No

Yes

Yes

No

Game Controller

Yes

No

No

No

No

No

No

Microphone

Yes

No

Yes

No

Yes

No

No

Touch Screen

Yes

Yes

Yes

Yes

No

Yes

Yes

Webcam

Yes

No

No

No

No

No

No

Eye Tracker

Yes

No

No

No

No

No

No

Laser Pointer

Yes

Yes

Yes

No

No

Yes

Yes

Accessibility Device and Access Options Accessible content should be furnished to the customers from the aggregator platform. The publishers who furnish the content material will comply with the suggestions for integration with assistive technological know-how tools. The

140 Advances in Data Science-Driven Technologies

Bhagvan Kommadi

required tags and labels for accessibility should be treated at some point in the facts switch and aggregation. The content material should be examined for assistive equipment through the usage of the accessibility guidelines. The repositories which hold the content material should make sure of the accessibility requirements. The repository property should be cataloged and be searchable by using the user and the reader. Designing the content material repository website is very necessary from a low imaginative and prescient user and cognitive overload issues. We have been discussing key elements, such as menus, navigation, content, web page structure, color, contrast, forms, images, keyboard accessibility, links, and language. Now let us focus on the typography and different key factors. The internet web pages reflow should not have a loss of content. Content needs to be resizable. The person can override the textual content spacing and have the content material on hand for the reader. The horizontal scrolling must be limited. Regarding mobile content material and accessibility, let us look at iPhone accessibility alternatives available. iPhone’s accessibility selections Table 3 for imaginative and prescient and speech are presented below: Table 3. Accessibility Device & Access Options Value/Access

Yes

No

Voice Over

On

Off

Zoom

On

Off

Magnifier

On

Off

Invert Colors

Yes

No

Greyscale

Yes

No

Speech

Yes

No

Larger Text

On

Off

Bold Text

Yes

No

Button Shapes

Yes

No

Vision and Speech Accessibility Options The selections for speech and interplay Table 4 of iPhone accessibility are presented below: Table 4. Vision & Speech options Value/Access

Yes

No

Larger Text

On

Off

Accessibility

Advances in Data Science-Driven Technologies 141

(Table 4) cont.....

Value/Access

Yes

No

Bold Text

Yes

No

Button Shapes

Yes

No

Increase Contrast

Yes

No

Reduce Motion

On

Off

On/Off Labels

Yes

No

Switch Control

On

Off

Assistive Touch

On

Off

Touch Accommodations

On

Off

Speech and Interaction Options The choices for interplay and listening Table 5 to iPhone accessibility are presented below: Table 5. Speech & Interaction Options Value/Access

Yes

No

Keyboard

Yes

No

Shake to Undo

On

Off

Vibration

On

Off

Call Audio Routing

Automatic

Manual

Home Button

Yes

No

Hearing Devices

Yes

No

LED Flash for Alerts

On

Off

Mono Audio

Yes

No

Direction

Left

Right

Interaction and Hearing Options The alternatives for the media and studying Table 6) of iPhone accessibility are presented below: Table 6. Interaction & Hearing Options Value/Access

Yes

No

Audio Volume Balance

Left

Right

Subtitles & Captioning

Yes

No

Audio Description

On

Off

142 Advances in Data Science-Driven Technologies

Bhagvan Kommadi

(Table 6) cont.....

Value/Access

Yes

No

Guided Access

On

Off

Accessibility Shortcut

On

Off

Media and Learning Options In mobile content, content material views in exclusive orientation should be restricted. Single-factor activation activities must be ensured to be canceled. The performance needs to be operable with the usage of a single pointer. User interface factors can be activated through movement and different mechanisms. In multi-media, accessibility is regarded by ensuring the audible content material is stated in the captions. Visual content material is stated in the audio of the multimedia content. The synchronized audio and video are furnished with media content. Synchronized captions for the video need to be provided. Audio should not be played on the load of the application. Playback of the media content material needs to be controlled. Media playback controls need to be mapped with shortcut keys in the application. The calendar controls need to be keyboard-accessible. Color has to no longer be the sole way to deliver the resolution of the calendar component. The information tables for calendar controls need to have header and facts cells. The non-modal calendars need to be rendered in line with the activating controls. The center of attention needs to go by activation and deactivation of the calendar controls. Dialogs need to have a desirable structure. The keyboard focal point needs to return from dialogs properly. The spawn dialogs from the hyperlinks have to exhibit the information. The non-modal dialogs need to be rendered in line with the controls which set off them. The dialogs need to be designed for closure by way of the keyboard. The modal dialogs need to have the keyboard focus. The dialogs which are activated need to have a focal point that goes accordingly. A dialog needs to have the title defined. The multi-column listing view controls need to have the rows selectable for the usage of the keyboard. The headers have to be sortable for the usage of the keys. The textual content options need to be supplied for the sortable headers. The web page tabs for the content material need to have a name and role. The menus need to be standalone, connected, and context-specific. The menus need to be connected to the fields. The menus need to have keys mapped, and programmatic focal points need to be provided. When the menus are closed, the keyboard focal point returns correctly. The menus need to be openable from the keyboard. The sub-menu objects need to be furnished with on-hand keys, and the

Accessibility

Advances in Data Science-Driven Technologies 143

sub-menu shape needs to be knowledgeable via help. The menus need to be rendered in line with the controls which are activating. DESIGNING FOR ACCESSIBILITY In one of the user journeys conferences, a well-known quote used to be made that ‘Accessibility is solved at the layout stage’. Designing for accessibility [5] is assembling the person's requirements. The user requirements deal with a couple of cultures, countries, kinds of users, specific demographic customers, and target users. Accessibility pursues visually impaired users and customers with varieties of impairments as nicely such as hearing, cognitive, mobility and motor-related problems. Accessibility is solved at the sketch stage. This is a phrase that Daniel Na and his crew heard over and over once more whilst attending a conference. The requirements of the customers need to be protected for the design of accessibility framework (Fig. 2). This consists of your goal users, customers' backyard of your goal demographic, customers with disabilities, and even customers from unique cultures and countries. Understanding these requirements is necessary for developing handy person experiences for them.

Fig. (2). Accessibility Framework.

144 Advances in Data Science-Driven Technologies

Bhagvan Kommadi

User requirements are gathered thru personal interviews. Different user requirements which tackle the person's journey are animation and effects, audio and video, color, controls, font, photos and icons, keyboard, layout, material, readability, shape, and time. Design concerns Table 7 imposed with the aid of the accessibility requirements and better person ride assist in growing UX wealthy products. Web pages are designed so that they can be tabbable with the keyboard for customers disabled and mouse utilization is a hassle for them. The content material and layouts h to be structured with the usage of HTML5. Shortcuts need to be furnished for convenient navigation of the user. Table 7. Design Concerns Context

Web Accessibility Improvement [6]

Content

Unique title and identifier for content and elements

Organization

Headings for content

Simplicity

Plain English is used

Visibility

Easy to read text

Descriptive

Links need to be more descriptive

Images

Alt text for images

Audio and Video

Text Alternatives for audio and video

Data

Tables are used

Color

Color contrast is important

Anti-patterns

usage of Text images

Animations, images, consequences make the net website online interactive and assist in branding the website. They can be perplexing and develop distractions to some of the internet site users. Flashing outcomes can motivate photosensitive epilepsy. These outcomes can result in headaches, nausea, and dizziness. Parallax and action results can make customers experience a vertigo-like effect. This can be due to the fact of vestibular sensitivity. Animations need to be for eventualities the place brief distances are simulated, and objects are shifting at lesser speeds relative to the display size. Users need to have the functionality to swap off and on some of these consequences at web page degree and world stage settings. Web Accessibility Improvements Typical questions which should be protected for animations, images, results are associated with seizure, dizziness, vertigo motion, distractions due to moving, blinking, and auto-updating. Videos and audio assist in growing higher person experience. Autoplaying is an alternative however it diverts the user interest and

Accessibility

Advances in Data Science-Driven Technologies 145

should be prevented throughout the layout phase. Videos must be averted enjoying in the background. Users ought to have settings to play, pause and quit the video. The content material like animations, images, movies and audio need to have transcripts, captions and subtitles to assist the customers accordingly. Blind customers may pick a transcript for movies and animations. Audio and movies should be designed to search quite a several elements associated with autoplay, controls, description, and hiding the content. Avoid facts in the download icons. Designing the internet website for specific organizations of customers helps in a higher user experience [7]. The age of the customers can fall into the young, center, and old categories. Color is an essential component in the format of web pages. It helps in branding and personal appreciation of the website. For blind users, coloration honestly no longer helps. Users with color blindness may additionally now not be in a position to make the distinction between red, green, yellow, and orange. Deuteranopia ailment can motivate blindness for inexperienced color. Users with this sickness discover pink as brown or yellow. They sense experience as beige. Typically guys have this disease. Protanopia sickness reasons blindness to pink color. For them, purple is perceived as darkish or black. Orange and inexperienced coloration appears like yellow. Men have this type of disease. Tritanopia disorder happens in guys and women. This sickness can purpose blindness to blue-yellow color. The blue color is perceived as inexperienced or teal. The yellow color appears like violet or grey. Color mixtures such as purple, green, yellow, and blue-pink need to be avoided. Internationalization and localization of content, images, icons, and net pages aspect in the shade particular to that united states and cultures. There is a distinction in japan and Asian cultures concerning the illustration of colorations to tremendous and bad trends. Patterns, icons, and textual content descriptions are used to deal with colorings and cultural differences. Icons that are crammed are higher than outlined ones. Color is used for highlighting the content material and exhibiting errors. For making the pages and shapes visually accessible, tooltips, thick borders, daring text, underlining, and italics can be used to exhibit errors. The contrast ratio of textual content to its returned floor is some other issue that needs to be minimal, from 4.5 to 1 per WCAG. Based on the font measurement of 19 or 24 pixels bold, the minimal modifications to three: 1. Customers with low vision, coloration blindness, worsening imagination, and prescient can study the textual contentbased totally on the contrasting factor. The different settings, like Windows High distinction mode, affect the plan of the pages with colors. The sketch elements to

146 Advances in Data Science-Driven Technologies

Bhagvan Kommadi

be regarded for dealing with colorings are barring the use of color, oversaturated colors, distinction factor, foreground, and, again, floor colors. Interactive content material, such as UI controls, assists customers in interacting with buttons, links, inputs, and other factors associated with the internet web and content. The measurement of the controls is a necessary aspect of the visibility and invocation of the controls. Customers with tremors and missing dexterity may have a hassle in pointing the controls. The control’s place and dimension is a layout aspect associated with clicking on the target. Users with widespread or adaptive pointing machines should have a solution for pointing and clicking the control. W3C’s Authoring Practices for Design Patterns [8] recommends reachable sketch patterns, including menus, modals, autocompletes, trees, and tab sets. Hovering needs to be avoided by designing for accessibility. Secondary movements in the menus or non-modal dialogs should be used alternatively of hover states. They can conceal the trigger. Secondary icon's coloration and distinction should be lightened. On hover, they should be darkened. Instead of white space, the data icon is higher to invoke a content material crammed hover. The function of controls performs a necessary function for contact display screen users. WCAG suggests the minimal dimension for the UI manipulation is 44 x 44 pixels. Google recommends 48 x 48 pixels. The minimal measurement has padding for the control. The sketch elements for deciding on controls are associated with size, positioning, padding, nesting, textual content visibility, and the kind of display (touch or web). Form fields should be labeled inner the label element. Placeholder textual content has to no longer be used for naming the label. Material is a very necessary format component for web page design. The appearance and experience of an aspect are associated with the material. Styling the hyperlink as a button or putting them after every different is an issue. Autocomplete menu is another aspect that can solve issues for keyboard or display reader users. Material layout elements are associated with users, behaviors, the aggregate of awesome behaviors, and the experience of the elements. Text readability can motivate troubles like tiring and taxing mentally. The vital elements are associated with sentence length, paragraph size, and language complexity. Content should be designed with headings, lists, menus, footers, and images. Headings assist in grouping and summarizing the information. The layout elements for readability are thinking at the back of the text, simplicity of the language, size of paragraphs, and utilization of the UI controls. Suggested line and

Accessibility

Advances in Data Science-Driven Technologies 147

paragraph spacing is atlas 1.5x for higher experience. Voice-over equipment assists blind customers in speaking their intent and description. The description of the content material and controls is placed as alt text. Live textual content should be used as a substitute for flattened. Avoid textual content in photographs all through design. Text hierarchy needs to be maintained. Text blocks of eighty characters should be used to assist customers with the slim discipline of vision. Bulleted lists are additionally recommended. Avoid underlining, italics, and capital-lettered text. Ensure spell take a look at is achieved for the textual content in the internet web pages. Idioms and Figure of speech should be avoided. The quantity of time spent by way of the user on a web page is a design factor. Session administration performs an essential position in time out of the classes and their expiry. Users can extend the session on a web page. During the design of the web pages, make certain the customers can ask for the favored channel for conversation or support. They should be given a realistic time for an animation to be finished. BEST PRACTICES The checklist below helps you to think about what to implement accessibility. Can app users perceive the content? Can application users use User Interface components and navigate the content? Can web app users understand the content? Can app users understand the interface? Is it consistent enough to avoid confusion? Can the content be consumed by a wide variety of web app user agents on the browsers? Does the web application work with assistive technology? Now let us start looking at how to implement accessibility after getting the answer to the above questions: Let us start from the keyboard. For customers who can't or pick out now not to use a mouse, keyboard navigation is their most important capability of achieving the whole lot on screen. This target audience consists of customers with motor impairments, such as Repetitive Stress Injury (RSI) or paralysis, as properly as display reader users. For an appropriate keyboarding ride purpose is to have a logical tab order and, without problems, a discernable center of attention style.

148 Advances in Data Science-Driven Technologies

Bhagvan Kommadi

Start with the aid of tabbing via your site. The order in which factors are targeted ought to purpose to comply with the DOM order. If you’re not sure which factors need to acquire focus, see the focus fundamentals for a refresher. The widely widespread rule of thumb is that any interaction with entering needs to be focusable and show a center of attention indicator (e.g., a center of attention ring). It is a frequent exercise to disable the center of attention patterns except by offering a choice via the usage of the outline: none in CSS, however, this is an anti-pattern. If a keyboard user can’t see what’s focused, they cannot interact with the page. If you should differentiate between mouse and keyboard focal points for styling, reflect on the consideration of adding a library like what-input. Custom interactive controls must intention to be focusable. If you use JavaScript to flip a element into a fancy dropdown, it will no longer robotically be inserted into the tab order. To make a customized manipulation focusable, provide a tabindex=”0”. Avoid controls with a tabindex > zero. These controls will bounce beforehand of the whole lot else in the tab order, regardless of their role in the DOM. This can be difficult for display reader customers as they tend to navigate the DOM linearly. Non-interactive content material (e.g., headings) has to keep away from being focusable. Sometimes builders add a tabindex to headings due to the fact they assume they are important. This is also an antipattern because it makes keyboard customers who can see the display screen much less efficient. For display reader users, the display reader will already announce these headings, so there’s no should make them focusable. If new content material is brought to the page, strive to make positive that the user’s focal point is directed to that content material so they can take motion on it. Beware of absolutely trapping center of attention at any point. Watch out for autocomplete widgets; the place keyboard focal point may also get stuck. Focus can be quickly trapped in unique situations, such as showing a modal, when you do not prefer the user interacting with the relaxation of the web page - however you need to purpose to supply a keyboard-accessible technique of escaping the modal as well. Just due to the fact something is focusable doesn’t suggest it is usable. If you have constructed a customized control, the purpose for a user to be capable to attain all of its performance is the usage of solely the keyboard. You should not neglect offscreen content. Many websites have offscreen content material that is currently in the DOM but no longer visible, for example, hyperlinks inner a responsive drawer menu or a button internal, a modal window that has but to be displayed. Leaving these factors in the DOM can lead to a puzzling keyboarding experience, specifically for display readers, which will announce the offscreen content material as if it’s a section of the page.

Accessibility

Advances in Data Science-Driven Technologies 149

You can try using a display reader. Improving widespread keyboard assist lays some groundwork for the subsequent step, which is to take a look at the web page for suited labeling and semantics and any obstructions to display screen reader navigation. If you are unfamiliar with how semantic markup receives interpreted with the aid of assistive technology. Check all photographs for ideal alt text. The exception to this exercise is when photos are chiefly for presentation functions and are now not quintessential portions of the content. To signify that a picture must be shipped via a display screen reader, you need to set the cost of the alt attribute to an empty string, e.g., alt=””. Check all controls for a label. Customized controls may additionally require the use of aria-label or aria-labeled. Check all customized controls for a fabulous function and any required ARIA attributes that confer their state. For example, a customized checkbox will need a role=” checkbox” and aria-checked=” true false” to right deliver its state. The glide of facts must make sense. Because display screen readers navigate the web page in DOM order, if you have used CSS to visually reposition elements, they may additionally be introduced in a nonsensical sequence. Aim to guide a display screen reader’s navigation to all content material on the page. Avoid letting any sections of the website be completely hidden or blocked from display reader access. If content material must be hidden from a display screen reader, for instance, if it’s offscreen or simply presentational, make certain that content material is set to aria-hidden=” true”. Familiarity with even one display reader goes a long way. Though it may appear daunting to research a display reader, they’re fairly easy to pick out. In general, most builders can get via with simply a few easy key commands. If you’re on a Mac, test out this video on the use of VoiceOver, the display reader that comes with Mac OS. If you’re on a PC, take a look at this video on the usage of NVDA, a donation-supported, open-supply display reader for Windows. It’s necessary to apprehend that ARIA can solely affect the semantics of an element; it has no impact on the conduct of the element. While you can make a thing hidden to display screen readers with aria-hidden=” true”, that no longer exchanges the center of attention conduct for that element. For offscreen interactive content, you will frequently mix aria-hidden=” true” and tabindex=”-1” to make certain it is eliminated from the keyboard flow. The proposed inert attribute objectives make this less difficult by combining the conduct of each attribute.

150 Advances in Data Science-Driven Technologies

Bhagvan Kommadi

Interactive factors like hyperlinks and buttons need to point out their motive and state. Providing visible pointers about what a manipulation will do, helps human beings function and navigate your site. These recommendations are referred to as affordances. Providing affordances makes it feasible for human beings to use your website on a large range of devices. Interactive elements, like hyperlinks and buttons, must be distinguishable from non-interactive elements. It is challenging for customers to navigate a website or app when they can't inform if an issue is clickable. There are many legitimate techniques to accomplish this goal. One frequent exercise is underlining hyperlinks to differentiate them from their surrounding text. Similar to the center of attention requirement, interactive factors like hyperlinks and buttons require a hover for mouse users, so they understand if they are hovering over something clickable. However, the interactive component nonetheless needs to be distinguishable on its own. Relying on a hover on my own to point out that clickable factors no longer assist contact display screen devices. You need to use headings and landmarks. Headings and landmark factors imbue your web page with semantic structure, and noticeably make bigger the navigating effectivity of assistive technological know-how users. Many display screen reader customers document that when they first land on an unfamiliar page, they generally attempt to navigate with the aid of headings. Similarly, display screen readers additionally provide the capacity to leap to necessary landmarks like and . For these motives, it’s essential to think about how the shape of your web page can be used to inform the user’s experience. Make desirable use of the h1-h6 hierarchy. Think of headings as equipment to create and define your page. Don’t count on the built-in styling of headings; instead, think about all headings as if they had been the equal dimension and use the semantically excellent degree for primary, secondary, and tertiary content. Then use CSS to make the headings fit your design. Use landmark factors and roles so customers can omit repetitive content. Many assistive applied sciences grant shortcuts to leap to unique components of the page, such as those described by way of or elements. These factors have implicit landmark roles. You can additionally use the ARIA position attribute to explicitly outline areas on the page, e.g., . See the information on headings and landmarks for greater examples. Avoid role=”application” until you have prior experience working with it. The software landmark position will inform assistive technological know-how to disable its shortcuts and bypass via all key presses to the page. This skill that the keys display reader customers usually use to go around the web page will no longer work, and you will put all keyboard dealing with yourself into effect.

Accessibility

Advances in Data Science-Driven Technologies 151

Quickly evaluate headings and landmarks with a display screen reader. Screen readers like VoiceOver and NVDA supply a context menu for skipping to essential areas on the page. If you’re doing an accessibility check, you can use these menus to get a speedy overview of the web page and decide if heading degrees are splendid and which landmarks are in use. You need to look at automating the process. Manually checking out a website for accessibility can be tedious and error-prone. Eventually, you’ll choose to automate the method as an awful lot as possible. This can be achieved via the use of browser extensions, and command-line accessibility take a look at suites. AXE and WAVE Browser extensions are simply two reachable choices and can be a beneficial addition to any guide check technique as they can rapidly choose up on refined problems like failing distinction ratios and lacking ARIA attributes. If you select to do matters from the command line, axe-cli presents the identical performance as the aXe browser extension, however, can be without difficulty run from your terminal. To keep away from regressions, mainly in a non-stop integration environment, include a library like axe-core into your computerized take a look at suite. axecore is the equal engine that powers the axe chrome extension, however in an easy-to-run command-line utility. You need to check for accessibility tools if you are using a tool or software. Some examples encompass protractor-accessibility-plugin for Angular, and ally suite for Polymer and Web Components. Take benefit of handy equipment on every occasion feasible to keep away from reinventing the wheel. DIGITAL ACCESSIBILITY Disabled customers should have the functionality to navigate and view the internet web applications, documents, and cell applications. This functionality is referred to as Digital accessibility. The incapacity can be associated with the eye, ear, body, and intelligence of the user. Many types of equipment exist for disabled customers to analyze the content material on exceptional channels. Users with impaired motor troubles use extraordinary tools for analyzing digital content. They can't use the traditional keyboard and mouse. They will use a display reader and a sip-and-puff swap for input [9]. Many of the websites make it difficult for disabled customers to access the content material, making it fantastic for them. Best practices advise the internet person interface factors to be designed well. Universal high-quality practices are used for designing websites to meet the requirements of more than one kind of user. Alt Text tags for photos and portraits assist internet customers who are

152 Advances in Data Science-Driven Technologies

Bhagvan Kommadi

listening to the impaired. Similarly, video content material having captions assist impaired users. Search engines index the websites which are accessible. Welldesigned websites for accessibility assist in making the websites searchable. Users with disabilities can search shortly to browse integral internet web content. The content material designed for the internet web would possibly have a specific user’s trip on the display reader for a disabled user. Screen readers become aware of well-known HTML elements. Standard factors, such as header, menu, body, and footer, are recognized by using the display reader. Keyboard navigation should be enabled due to the fact disabled customers would possibly no longer be capable of using the mouse or browsing the content. Disabled customers with muscle management issues and who have lost arms can use the keyboard for navigation. User moves should be designed to retain in thought the visible flow, navigation path from left to right, pinnacle to backside looking, and navigating the menu. The different movements that should be viewed for layout are clicking on the centered link, shopping the footer, and closing the UI home windows and tabs to separate views. UI factors, such as buttons, links, shape fields, and records pickers, are the correct candidates for designing for accessibility. ALT attribute for img is shown below.

The input tab needs to be taken care of for the disabled user. Focusable factors and indications in an internet website need to be recognized in the design. CSS factors are used to format the visible factors on a net page. The attributes for the description of URLs should be significant and descriptive. For content material like photos and videos, ALT attributes can have descriptive and significant phrases. ARIA label helps the display screen reader in analyzing the name to motion textual content on the button. They are used to override the HTML labels for designing an internet website for accessibility. A sample for a href tag is presented below with an ARIA label. Sign up The HTML labels should be seen, and the structure should not have placeholder text. The varieties should be designed in a single-column format. The borders for the textual content area, content fields, and drop-down menus should be considered for designing the internet web types for accessibility. HTML information tables should have a CSS for layout. HTML tables are generally used for tabular data. Page Layouts should have CSS for designing the pages for disabled users. Each area on the web page needs to have descriptive headings.

Accessibility

Advances in Data Science-Driven Technologies 153

Semantic markup should exist for headings, paragraphs, lists, and quotes. The web page format needs to be designed to aspect in the left-aligned textual content for convenient scanning of the content. The textual content headlines should be centered. Text alignment should be chosen as a theme for the web page design. The web page fonts should be chosen for web page design. Basic fonts should be used. For disabled customers with a visible disability, Sans-serif fonts are used in the net design. Picking the proper font measurement is important. Font sizes larger than 12 are recommended. Choosing relative gadgets for font dimensions helps make the content material readable without difficulty. Animations such as blinking and transferring textual content are now not recommended. Color scheme and distinction should be viewed for internet web design. According to WCAG (Web Content Accessibility Guidelines), the 4.5:1 distinction ratio can be used for text. The different hints are a 21:1 ratio for black textual content on a white background and 4.5:1 for grey textual content on a white background. The web page factors such as shapes or icons should be blended with a color. Web site layout needs planning for making sure accessibility for disabled users. ACCESSIBILITY PROJECT LIFECYCLE Many customers are unable to get the right of entry to the data on the net, which is inflicting a dichotomy. Websites should be designed for accessibility for human beings with disabilities. Accessibility helps a person search the internet website and the usage of the information. HTML5 is the language used for designing and growing the website. Mobile, PDA, and Televisions Browsers on computer systems get admission to the HTML5 content. The web browser is properly recognized for getting access to HTML5 net-based content. Users with impaired eyesight may have problems with content, contrast, shade, and pictures on the internet web pages. Screen readers are used for remodeling the internet web page content material to speech. Jaws, Window-Eye, NVDA, Serotek System Access, Apple VoiceOver, ORCA, BRLTTY, Emacspeak, WebAnywhere, Spoken Web, ChromeVox, and ChromeVis are the nicely-acknowledged display reader. Blind Web customers can use a display screen reader to rework the textual content to a shape in which the person can process the information. Some display screen readers convert the textual content to speech, while others use the refreshable braille technique for display. The display readers are primarily based on the piezo impact in which the crystals amplify on particular voltage levels. Fingers are used by way of visually impaired customers to study the text. An accessibility project lifecycle (Fig. 3) for making an internet website reachable consists of a couple of phases, such as planning, analysis, design, development, checking out and maintenance. The project phases consist of net web page design, outlining of the structure, internet web page template creation, integration of templates, content

154 Advances in Data Science-Driven Technologies

Bhagvan Kommadi

material addition, and internet website launch. Design and user interface conceptualization [10] is an essential segment of the accessibility undertaking lifecycle. Web applications require accessibility to become popular and attract a high number of readers. If accessibility is not considered very early, refactoring at later periods challenges planning problems.

Fig. 3. Accessibility Lifecycle

Website design has achieved the usage of wireframes, and accessibility elements should be brought into the design. The selections made for accessibility should be shared with designers, builders, and testers. Accessibility should be built-in in the course of improvement, and the unit trying out needs to take care of the requirements. The QA crew validates the accessibility requirements and ensures the internet website performs properly with assistive science and web browsers. For the accessibility project, metrics and monitoring of the goals are necessary. The intention is to have an internet website on hand for customers with disabilities. Accessibility projects will have short and long-term goals for managing the product aspects and accessibility requirements. The metrics measure the accessibility improvements. They pick out the problems with internet web design, defects, and prioritization issues. Accessibility is a tough design problem. Expertise is required for managing accessibility requirements in the code. The problems commonly are sketch issues for inaccessible web page elements. They are associated with WCAG's two tenet violations. Accessibility architects and builders should design and boost the internet website based totally on the guidelines.

Accessibility

Advances in Data Science-Driven Technologies 155

The development and Testing [11] crew should have the trying-out equipment for unit checking and user testing. Different browsers have extensions and plugins for trying out accessibility requirements. The trying out equipment for accessibility is deliberate very early for net website releases. The equipment used for checking out should be checked for no longer supplying false positives. QA group validates the requirements of the use of equipment and guide testing. Accessibility project testing [12] is like any QA project. The testers and the builders should comply with the accessibility recommendations and requirements. Both groups should have equal equipment for checking and retaining the internet website for accessibility. The internet web Forms should not be designed with the nesting of structure factors and links. Placeholder values to the label should not be used. Accessible utilization of time-based periods and timed responses are to be viewed for the structure design. CAPTCHAs should be on hand for each visually and audibly. The checkboxes and radio buttons should be placed to the left of the labels. The internet web page factors with extra than one label need to be designed properly for excellent rendering. The error messages need to be highlighted before establishing the shape after submitting. The structure issue hierarchy needs to be highlighted in a textual manner. The structure discipline constraints and blunders should be placed in the corresponding fields. The structure of discipline labels should be unique. The structure fields need to be laid out in an intuitive order. The choice factors in big lists need to be grouped. The radio button groups need to be designed well. The rich edit entry fields should be accessible without delay. The frequent enter fields are to be designed for auto-completion. The fields must use well-known autocomplete values. The structure of instructive textual content needs to be positioned at the beginning of the form. The seen textual content label for the manipulation should be brought in the handy title of the control. The error facts for the fields should be truly indicated. The structure needs to have a steady implementation of error and alert mechanisms. A valid label for the structure fields should be given. Audio cues are to be furnished as alternatives. Error prevention for felony commitments, monetary transactions, taking a look at responses, and statistics adjustments should be treated properly. Field units for corporations of structure controls need to be designed well. The recommendations for error messages should be indicated when known. Valid, concise, and significant choice textual content ought to be furnished for photograph buttons. The visible labels and directions for app users to enter must be designed well.

156 Advances in Data Science-Driven Technologies

Bhagvan Kommadi

The internet web page hyperlinks should have the choice of textual content for photograph links. The hyperlink textual content must be significant for situations. Link textual content needs to be significant in the context. The internet web links should be grouped. The identical hyperlink textual content for hyperlinks needs to be prevented with distinct targets. The internet web page shape must no longer use implicit headings. Heading factors should be averted if now not necessary. Long costs should be treated with the use of a block quote. Proper spacing needs to be furnished for complicated textual content elements. The internet web page content needs to be designed for assistive equipment such as display readers and others. The internet web page content material needs to be hidden from internet customers who cannot be rendered through assistive technology. You should have the heading stage fit the heading's visible level. The headings and labels should be descriptive and unique. The markup archives must have well-formed elements. The web page studying order should coincide with the center of attention order of the internet web page. Title components have to be furnished for the internet web pages. Proper markup has to be used to mark emphasized or one-of-a-kind text. The acceptable citation markup should be used. The analyzing order of content material ought to be logical. You should keep away from the use of ASCII layouts. Page titles should be informative and context-sensitive. Keyboard gets entry needs to be supplied for scrollable content. PLANNING FOR ACCESSIBILITY Accessibility makes the software program beneficial for many users. It is about making sure the websites and the computing device software program help customers who have disabilities, mobile customers, and sluggish internet-based users. Accessibility gives equal possibilities and helps to manage exceptional organizations of customers in a better way. Customers with distinct disabilities are given equal rights to get the right of entry to ordinary users. Accessible software program helps to make the internet website extra search engine optimization friendly [13]. The company’s which presents a handy internet web page, has a suitable branding image. The customers with disabilities are visually impaired internet users, customers with motor characteristic issues, and listening-to-impaired users. Blind customers may use display readers to access the content material at one-of-a-kind zoom factors. Visually impaired customers may have low degree vision, blindness, and color blindness. Motor feature hassle dealing with customers would possibly use keyboard and different points which are non-mouse. These customers would possibly have paralysis or non-working limbs. Their fingers and legs may be

Accessibility

Advances in Data Science-Driven Technologies 157

susceptible due to neurological disorders. Deaf customers may use captions and different content material options for voice and video. The different crew of customers may have disabilities that are associated with cognitive impairment and situations in thinking/memorizing. They would possibly have illnesses like intellectual illness. These ailments would possibly be depression, schizophrenia, dyslexia, and interest deficit hyperactivity disorder Users who have these disabilities should get entry to digital content material and websites. They will be successful to navigate, enter and study the content material with the use of specific assistive technologies. Americans with Disabilities Act of 1990 (ADA) regulation stipulates that websites be available to disable users. Web content material accessibility pointers (WCAG) are published as a standard. These tips are associated with the perceivability, operability, understandability, and robustness of the net content. WCAG compliance stages are A, AA, and AAA. Planning for internet website accessibility [14] should begin early in the software program launch process. The group needs to have accessibility professionals proper from the beginning of the project. They help to cut down remodeling and reduce the problems for the duration of the checking-out stages. Training the group of builders and testers on accessibility is very necessary. The mission layout needs to factor in time for designing [15], development, testing, fixing, and retesting. It would help if you had a requirements board to figure out which websites [16], internet web pages, content material administration systems, and visible designs should be developed. This board decides the precedence for the requirements. Any accessibility issue should be cautiously reviewed using the board for accessibility requirements. Implementation of these requirements should take place in phases or sprints. The team can have a roadmap for future requirements, which enhances the capability of the application. Accessibility is no longer a onetime fix. It is a non-stop procedure, and as new internet web pages are developed, the accessibility requirements come in. Testing and fixing problems is another phase of the project. The match handles must now not have navigation and structure submission triggers. Meta redirects should be avoided. The pages should not refresh automatically. The internet web pages should have a consistent navigation shape. The voice can be the mode for accessing the content. A webpage needs to be locatable in a couple of approaches from a team of pages. The previous repetitive content material needs to be skippable. A way should be furnished to omit the previous repetitive hyperlinks with the aid of the use of seen links.

158 Advances in Data Science-Driven Technologies

Bhagvan Kommadi

To make certain Keyboard Accessibility, inactive factors should be averted in the focal point order. The gadget-structured tournament handlers need to now not be used alone. The title attribute should not be used alone. Accelerator and shortcut keys must be special and must be regular throughout the net pages. The energetic factors should have a keyboard focal point and can be activated by using the keyboard. The factors which are read-only and editable should have the keyboard focus. The JavaScript primarily based net elements should be keyboard accessible. The personal key shortcuts have to be ensured for reconfiguration and deactivation. The content material which is hovering or focused can be closed with the aid of the person if necessary. The center of attention order of the interactive factors should be logical. Documentation needs to be supplied for nonwidespread keys which are used for access. ACCESSIBILITY PLATFORM The reader will comprehend the platform modules and aspects of the typical internet website accessibility platform. We will be focusing on key elements which affect accessibility, such as frames, information tables, charts, graphs, bushes and outlines, web page tabs, dialogs, calendar controls, animations, dynamic content, authoring tools, sketch tables, mobile, multi-media, multi-media manage playback, and typography [17]. Computers have started from DOS-running gadgets that used to be textual content-based. The display had characters, and the cursor once displayed the role on the screen. For accessibility, the content material can be studied from the display screen, and moves can be intercepted earlier than dispatched to the screen. The content material can be modified to unique zoom patterns or choice structures. Table 8. Platform Features Disability

Assistive Technologies

Visual Disabilities

Screen readers, Screen Magnifiers, Color blindness, Audio signals, Color contrast, Audio Descriptions

Mobility Disability

Keyboard users, voice recognition

Psychological Issues

text to speech, screen overlays

Cognitive Problems

Text to Speech, Voice recognition, color overlays

Hearing Disability

Captions, Visual Cues

ESL

Text to Speech

Retirees

Screen Readers, Text to Speech, Screen Magnifiers

Accessibility

Advances in Data Science-Driven Technologies 159

Disability - Assistive Technology From DOS to different working structures like Mac, Windows, and Linux, interfaces are extra graphical. Images, movies and different content materials are treated in these running machines for usability [17]. The graphical engine handles the API calls for managing text, images, formats, actions, and activities from the interface. An off-display screen model can be developed from the API objects, calls, occasions, and moves which can be used by using the display readers and display screen magnifiers [18]. The offscreen model is generally a set of screenreaders, voice dictation software, and speaking phrase processors with phrase prediction. The offscreen model should have the interface context, objects, moves, and the running machine's precise native methods. The position and country of the interface objects are an additional section of the model. The model is based on the working gadget's precise facets, and the accessibility strategies to cope with these facets evolve with the running system. The model needs to have content material alignment, white spacing, and formatting rules. The internet web browsers guide the accessibility APIs, such as Microsoft Active Accessibility (MSAA), Microsoft UI Automation (UIA), and UI Automation Express (UIA). Now let us look at the Accessibility platform, which can assist customers in lights and administrative center places who have issues in reading, writing, and scanning content material from distinctive content material kinds and applications. The platform will have the enterprise requirements and the standards for accessibility. The best practices will be running devices precisely and based on the content material types. The concepts, auditing checklists, legal guidelines, and strategies are the phase of the accessibility science platform [19]. They can be used for distinct system kinds, such as mobile, desktop, laptop computers, and different machine types. The platform offers reviews for the below: popular particular compliance, violations for exceptional practices, Accessibility repute throughout the content, compliance rates, severity ratings, noticeability ratings, tractability ratings and recommendations for issues The problems and reviews are introduced primarily based on the following guidelines: WCAG 2.0 [14] WCAG 2.1

160 Advances in Data Science-Driven Technologies

Bhagvan Kommadi

AA AAA Section 508 CVAA The platform handles exclusive content material types, controls, enter indicators, alerts, and alerts. The platform will guide cognitive accessibility. Cognitive accessibility is associated with reminiscence issues, the pace of processing data problems, enterprise, and coordination issues. We have discussed key elements such as menus, navigation, content, web page structure, color, contrast, forms, images, keyboard accessibility, links, and language [12]. Now let us look at the authoring tools. Authoring equipment should have aspects to produce content material that follows WCAG's two guidelines. The equipment should have skills for exporting to PDF and different UA-conformant documents. During the conversion of the formats, the equipment should hold the key statistics from the content. Accessibility particular templates for this equipment should observe the precise hints and standards. The internet website will have frames and iframes in the web page design. These frames need to have the textual content set right. The titles should be significant for the frames. The sizing of the frames has to be absolute. The frames should have unique source markups. The different internet website controls like format tables should be reachable conformant. The diagram tables should have absolute sizing The structural markup must now not be there in the design tables. The user wishes to be detailed, and these tables ought to be linearized. The future is in having accessibility platforms developed as software to help in the implementation of accessibility in software projects. Research GAPS The open research issues, research trends and future research directions can be categorized into different areas like standards/guidelines, website design phase, website development phase, and accessibility planning phases. The typical challenges or gaps are related to awareness, lack of skilled resources, availability of resources, and lack of accessibility implementation knowledge/courses. During the planning phase, different obstacles to implementing accessibility are faced. This might be because of a lack of initiative, training, and coaching. Government [20 - 23] and disabled support organizations are taking the initiative to design web applications for accessibility. Testing is another area that requires resources to test

Accessibility

Advances in Data Science-Driven Technologies 161

the web application for accessibility [24 - 27]. Accessibility research needs support and funding from the government to coach and train the resources to implement government websites for accessibility [28, 29]. Universities and educational institutions need to be encouraged to focus on accessibility research to train students in accessibility design and development. CONCLUSION In this chapter, we looked at different elements and factors of accessibility from a design perspective. Features of the accessibility platform were presented in detail, and the potential is in the platform to become a full-fledged commercial software for enterprises. Accessibility-related tools need to use for evaluation [23]. New frameworks need to be developed in the future to add accessibility checks in the browsers like WCAG Inspector for Firefox. Browser extensions need to be developed in the future to visualize accessibility-related concerns. New evolving technologies are planning to incorporate accessibility. The goal is to ensure the disabled or users with special needs are getting access to the new technological devices. The new networks like NGN and its associated services are planning to add accessibility to the services. CONSENT FOR PUBLICATION Not applicable. CONFLICT OF INTEREST The author declares no conflict of interest, financial or otherwise. ACKNOWLEDGEMENT Declared none. REFERENCES [1]

Microsoft accessibility. Microsoft Website, 2014.

[2]

Accessibility Testing, Accessibility Testing. Section508 Website..

[3]

Agile Accessibility Handbook, A Practical Guide to Accessible Software Development At Scale. Dylan Barrell. Amplify Publishing, 2021, pp. 10-50.

[4]

Inclusive Design for a Digital World: Designing with Accessibility in Mind (Design Thinking). 1st ed. Edition. Regine M. Gilbert. Apress, 2019, pp. 30-70.

[5]

Web Accessibility, Web Standards and Regulatory Compliance.

[6]

Richard Rutter. Apress, 2006, pp. 110-150.

[7]

"A Web for Everyone: Designing Accessible User Experiences", Sarah Horton & Whitney Quesenbery. Rosenfield 2014. pp 60 -90.

162 Advances in Data Science-Driven Technologies

Bhagvan Kommadi

[8]

Whitney Quesenbery. Rosenfield, 2014, pp. 60-90.

[9]

Building For Everyone, Expand Your Market with Design Practices From..

[10]

Google’s Product Inclusion Team, Annie Jean-Baptiste. Wiley, 2020, pp. 101-150.

[11]

Design for Real Life. Eric Meyer & Sara Wachter-Boettcher. A Book Apart, 2016, pp. 41-80.

[12]

Website Accessibility for Business, Tom Lavis. Huxley, 2021, pp. 60-98.

[13]

Access Technology for Blind and Low Vision Accessibility, Siu Yue-Ting. Ike Presley, 2019, pp. 1256.

[14]

Thinking About Web Accessibility H Robert King., 2019, pp. 18-46.>

[15]

Practical Web Inclusion and Accessibility, Ashley Firth. Apress, 2019, pp. 21-61.

[16]

H. Abuaddous, M.Z. Jali, and N. Basir, "Study of the accessibility diagnosis on the public higher institutions websites in Malaysia", In: Proceedings of the 4th International Conference on Computing and Informatics (ICOCI 2013).Sarawak, Malaysia, Universiti Utara., 2013, pp. 122-127.

[17]

H.S. Al-Khalifa, "WCAG 2.0 semi-automatic accessibility evaluation system: design and implementation", Computer and Information Science, vol. 5, no. 6, pp. 73-87, 2012. [http://dx.doi.org/10.5539/cis.v5n6p73]

[18]

M. Greeff, and P. Kotzé, "A lightweight methodology to improve web accessibility", In: Proceedings of 2009 Annual Research Conference of the South African Institute of Computer Scientists and Information Technologists (SAICSIT ’09) ACM Press: New York, NY, 2009, pp. 30-39. [http://dx.doi.org/10.1145/1632149.1632155]

[19]

K. Groves, A Challenge to Accessibility Testing Tool Vendors, 2013.http://www.karlgroves.com/2012/10/03/a-challenge-to-accessibility-testing-tool-vendors/

[20]

M.F. Theofanos, and J.G. Redish, "Bridging the gap", Interactions (N.Y.N.Y.), vol. 10, no. 6, pp. 3651, 2003. [http://dx.doi.org/10.1145/947226.947227]

[21]

S. Horton, S. and M. Quesenbery, A web for everyone: designing accessible user experience. Rosenfeld Media, 2014.

[22]

J. Nielsen, Why you only need to test with 5 users, 2000.https://www.nngroup.com/articles/why-yo-only-need-to-test-with-5-users/

[23]

S.L. Henry, Just ask: integrating accessibility throughout design, 2007.www.uiAccess.com/JustAsk/

[24]

K. Madathil, and J. Greenstein, "Synchronous remote usability testing: a new approach facilitated by virtual worlds", In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI ’11) ACM Press: New York, NY, 2011, pp. 2225-2234. [http://dx.doi.org/10.1145/1978942.1979267]

[25]

J. Nielsen, How many test users in a usability study?, 2012.http://www.nngroup.com/articles/howmany-test-users/

[26]

W3C, http://www.w3.org/WAI/eval/users.html

[27]

J. Lazar, B. Wentz, and A. Almalhem, "A longitudinal study of state government homepage accessibility in Maryland and the role of web page templates for improving accessibility", Gov. Inf. Q., vol. 30, no. 3, pp. 289-299, 2013. [http://dx.doi.org/10.1016/j.giq.2013.03.003]

[28]

S. Krug, A common sense approach to web usability. 2nd ed. New Riders Publishing: Berkeley, California, 2014.

[29]

R. Lopes, K. Isacker, and L. Carricco, "Redefining assumptions: accessibility and its stakeholders", In: Proceedings of the 12th international conference on Computers helping people with special needs:

Accessibility

Advances in Data Science-Driven Technologies 163

Part I (ICCHP'10). Klaus Miesenberger, Joachim Klaus, Wolfgang Zagler, and Arthur Karshmer (Eds.).Springer-Verlag, Berlin, Heidelberg, 2010, pp. 561-568. [http://dx.doi.org/10.1007/978-3-642-14097-6_90]

164

Advances in Data Science-Driven Technologies, 2023, 164-182

CHAPTER 8

Elderly and Visually Impaired People Mobility in Home Environment Using Adhesive Tactile Walking Surface Indicators Vijaya Prakash R.1,* and Srinath Taduri1 1

Department of Computer Science and Engineering, SR Engineering College, Warangal, India Abstract: Numerous health problems, particularly those involving the eyes, are associated with advancing age. It is difficult to live a normal life when you're blind. Visually impaired people face navigational difficulties both inside and outside of an environment, particularly those who are blind because of ageing. Numerous tools are available in the outdoor environment, such as pavement paths and kerbs. These, on the other hand, are ceramic, concrete, or metallic in nature, and once installed, their alignment cannot be altered. As a result, there is a need for adhesive-based tactile that is easily replaceable to meet the needs of the occupants of the house. The purpose of this paper is to design and develop various types of tactile using Thermoplastic Polyurethane (TPU) material and a 3D printer. These tiles include a Warning tile, a Straight tile, a Turning tile, and a Junction tile with surface indicators; elderly people can easily navigate their homes with the help of these tiles.

Keywords: 3D printer, Haptic Design, Junction Tile, Straight Tile, Surface Indicators, Tactile, Tactile surface indicators, Thermoplastic Polyurethane (TPU), Turning Tile, Visual impaired people, Warning Tile. INTRODUCTION According to the WHO [1, 2], there are 1.3 billion visually impaired people in the world, with 36 million of them being completely blind. Many blind people in the world reside in countries that are still developing [3]. This problem also affects the elderly, with the number of blind people over 65 growing at a rate of up to 2 million per decade, significantly faster than the overall population of blind people [3]. Individuals frequently rely on vision to determine their position and orientation in their environment, as well as to identify a variety of environmental components, as well as their distribution and relative placement. Corresponding author Vijaya Prakash R.: Department of Computer Science and Engineering, SR Engineering College, Warangal, India; E-mail: [email protected]

*

Manoj Kumar M V, Immanuel Azaad Moonesar R.D., Ananth Rao, Pradeep N, Annappa, Sandeep Kautish and Vijayakumar Varadarajan (Eds.) All rights reserved-© 2023 Bentham Science Publishers

People Mobility

Advances in Data Science-Driven Technologies 165

“Orientation” and “wayfinding” are common terms for these activities, while “mobility” refers to the ability to detect and avoid potential hazards in the immediate environment. Absent vision makes these activities more difficult, necessitating a determined effort to incorporate the experiences of other sense modalities, such as smells or sounds, into the activity. When individuals reach a certain age, their health begins to deteriorate, resulting in a variety of tied to the user's age age-related conditions, such as limb numbness, diabetes, hypertension, and hypotension. Proliferative diabetic retinopathy (neovascularization), Retinitis pigmentosa (pigmentary retinopathy), and cataracts all affect vision [4]. Certain of these ocular disorders can be treated surgically. Certain ocular disorders, on the other hand, are incurable and cannot be treated surgically. Normal life can be extremely difficult for elderly people with incurable ophthalmic disorders [5]. A tool for safe travel within the home environment is required in this situation [6]. Pavement pathways and kerbs are examples of tools that can be found in the real world [6 - 8]. Nearly every railway station in major cities now has tactile surface indicators (TSIs) for visually impaired passengers. Numerous blind individuals claim to use these TSIs when walking alone across railway platforms [9]. Ceramics can be replaced with concrete or metal, but once it's in place, it can't be moved. However, the furniture and other equipment in the house can be rearranged to better suit the occupants' requirements. More than 60% of Indians, according to official Figures, rent their homes. Changing residences will cause changes to their internal organization. As a result, there's a market for tactile adhesive tiles that can be swapped out and reused for visually impaired users. To satisfy the requirements of those who are blind, there is a demand for adhesive tactile tiles that can be changed and reused. Many navigation systems are available, but each has its own set of limitations in terms of mobility [9, 10]. In this context, the current effort focuses on establishing a more effective and efficient navigation system for visually impaired senior persons. This technology employs an adhesive tactile with bumps, which individuals of all ages can easily comprehend and identify, allowing them to explore their environment. The section that follows in the article is laid out as follows. The second section examines the most relevant works in the field. Within this section, we will talk about the system's architecture. Section 4 details the research conducted, while Section 5 offers some conclusions and suggestions for future endeavors.

166 Advances in Data Science-Driven Technologies

Prakash R. and Taduri

RELATED WORK For navigation in the outdoors, tactile walking surface indicators are frequently used. Many devices have been created to help and guide people with visual impairments through their daily activities, both indoors and outdoors. However, they did not meet all technical and user requirements. People with visual impairments face many challenges when it comes to spatial cognition and perception, and numerous research areas have emerged to address these issues [9]. First, this section will discuss related work on tactile standards. The second type of mobility for the visually impaired is sensory-based mobility, followed by tactile-based mobility. A large part of early studies focused on how to design, install, and use detectable warning surfaces around the world. There are different types of standards evaluated for tactile. The specification and standards for tactile warnings on the entire curb ramp walking surface have been discussed in the study [11]. There are warning textures, later referred to as detectable warnings, that are specified in the standard [12]. In 1988, the Australian/New Zealand Standard [13] required the specification of truncated dome warning surfaces, but the Australian Building Code did not require them until 1999. The ADAG (Americans with Disabilities Act Accessibility Guidelines) mandated, starting in 1991 [14], that blind travelers use truncated dome detectable warnings. The Accessibility Standard includes requirements for ADAAG truncated dome detectable warnings [15]. As a bonus, the texture and visual contrast requirements were identical to those of ADAAG. Tactile surfaces have been standardized by the European Union since 2002 [16]. Tactile walking surface indicators, a new ISO standard for assistive products for people who are blind or have low vision, were first used in 2004 and were finalized in 2012 [17]. For this paper, the goal is for the users to identify and prioritize the features of TWSIs that comply with the standards for guiding building environment designers. Various nations around the world are currently working on tactile surface solutions. By 2001, the Japanese Industrial Standard [10] had standardized the form and height of tactile walking surface indicators. As a result of these findings, it is possible to efficiently achieve “ease of recognition,” “ease of walking on tiles,” and “ease of recognizing transitions from bar tiles to dome tiles”. For people who are visually impaired, TWSIs can be extremely helpful. However, TWSIs should not only be used by the blind or visually impaired; they should be used by everyone, including the elderly and those with mobility issues. Tactile paving surfaces, on the other hand, are difficult to maintain and implement properly, and practitioners have had to deal with numerous issues. Specific recommendations on how to design pavements to fit the surrounding environment,

People Mobility

Advances in Data Science-Driven Technologies 167

for example, are frequently missing from standards [18]. People with visual impairments are classified into three groups based on how well adaptive visual systems improve their standard of living [6]. There are three types of vision enhancements: (1) vision improvement, (2) vision alternation, and (3) vision replacement. People with visual impairments have an opportunity to test out mobile and handheld assistive technologies by wearing them. Measurements of ultrasonic distance and haptic feedback have been used to develop additional white cane-related devices [19, 20]. Amedi et al. [21] used multiple infrared sensors to create electronic travel assistance with haptic and auditory output. Bharambe et al. [22] developed a sensory substitution system that looks like a hand device equipped with two ultrasonic sensors and three vibration motors. Yi et al. [23] developed a cane device that utilizes ultrasonic technology and incorporates haptic feedback and auditory guidance. Aymaz and Cavdar [24] have created an assistive headgear that uses ultrasonic distance measurements to look for potential hurdles in one's path to the wearer. Ultrasonic smart glass was developed by Agarwal et al. [25]. A low-cost smart walking stick with water and ultrasonic sensors was proposed by De Alwis et al. [26]. For visually impaired people, Elgendy et al. [27] created an indoor navigation system based on markers. The location begins when a QR code is printed on paper and placed in an interior area, and the destination can be identified using a mobile device. Petsiuk et al. [28] developed a navigation system for visually impaired individuals that makes use of an open-source, low-cost bracelet with a navigational support system based on ultrasound, assisting them in orienting themselves in their surroundings and avoiding obstacles while travelling. Table 1 includes a breakdown of the limitations of each system, as well as a more in-depth look at the technology. Table 1. Evaluation Of Reviewed Systems and their Limitations System Name

Smart-Cane [20]

Eye-Substitution [22]

Accuracy

N/A

N/A

Coverage

Limitation

Outdoor

If the water is only 0.5 inches deep, the water sensor will not work. The buzzer will continue to blare until the room is completely dry. To keep tabs on the situation, a power supply metre reading must be installed.

Outdoor

Uncomfortable to use because of the wood foundation and the holes in the Figures, as well as being carried around by the user. Utilization restricted solely to Android-powered devices

168 Advances in Data Science-Driven Technologies

Prakash R. and Taduri

(Table 1) cont.....

System Name

Accuracy

Coverage

Limitation

CASBlip [29]

80%

Indoor/ outdoor

The detection window is quite small. For image acquisition, you'll need a larger image sensor than a 1X64 CMOS one.

System of Cognitive Guidance [30]

N/A

Indoor

Enhancing the system's ability to perceive landmarks will be necessary. The reconstruction walking plane's stabilisation and registration should be improved.

To aid in navigation, an ultrasonic cane is being tested [31].

N/A

Indoor

It's only a simple object detector. A faint detection ring was heard. Objects that appear out of nowhere are not detected.

Automatically Adaptive Thresholding for Obstacle Avoidance [32]

N/A

Indoor

The accuracy of the Kinect depth image decreases with increasing distance from the sensor. After 2500 mm, the auto-adaptive threshold was unable to tell the difference between the floor and an object.

Haptic and Laser Rangefinder Obstacle Avoidance [33]

N/A

Indoor

It was difficult to pinpoint the exact location of obstacles and the angles at which they were to be overcome.

The FingerReader [34]

93.9%

Indoor/ outdoor

The audio feedback is responsive in real-time, but there is a noticeable pause between instructions.

Indoor

The information gathered is based on the volunteer's availability to provide it. There's a chance that no input will be received during the interval, so the service will fall short of its objective.

Indoor/ outdoor

The modules are very expensive. The impact of object evasion and navigation methods must be demonstrated through more extensive testing. However, the test's initial focus was on central scene object detection techniques.

Indoor/ outdoor

Obstacles higher than the level of the waist cannot be detected by the system. You won't be able to find any directions on this map. The area that can be reliably detected is quite small. It doesn't stand on its own.

Indoor

Every room would have to have sensors installed. Before anything else, the space needs to be measured and mapped out. The user is required to choose a destination in advance. It can't be used outside.

Blind People's Mobile Crowd-Assisted Navigation [35] Visually impaired individuals can benefit from wearing an ultrasonic assistive headset [24].

20.5%.

N/A

Use of ultrasonic technology for obstacle detection and recognition [36]

N/A

SUGAR system [37]

High Precision

Tactile indicators that are easily detectable and identifiable are required. Mizukami [10] conducted one of the most extensive studies ever conducted on the perceptibility of various types of tactile floor indicators. 68 blind people have assessed 21 different tactile floor indicators and 20 different types of attention fields in a training area. To find out how important haptic indicators are in non-

People Mobility

Advances in Data Science-Driven Technologies 169

visual navigation and what types of guiding surfaces people prefer, Saiganesh Swaminathan et al. [2] surveyed blind people and interviewed ten navigation and public accessibility experts. They also propose navtiles, which are low-cost, multimodal tactile surfaces, as a method for designing and exploring them. The blind and partially sighted in Yogyakarta, Indonesia, can benefit from based on the results of an investigation by Ardilson et al. [38] that looked at the current state of TSI (Tactile Surface Indicators) installations on secondary arterial roads. People with visual impairments are unable to use existing TGSI installations [38] because of poor connectivity, hazardous pathways, inaccurate installation, and inaccessible pedestrian walkways amenities. This has led to the discovery that TGSI equipment is both unusable and dangerous for those with visual impairments [38]. Most currently available technologies for visually impaired individuals are sensor-based, mobile, or wearable devices that are used indoors or outdoors. However, elderly individuals, particularly those living in developing countries, are unable to make more efficient and effective use of these devices. Therefore, an effective adhesive-based tactile walking surface indicator that can be used indoors or outdoors is required. TACTILE DESIGN METHODOLOGY Target Users Identification of stakeholders is a critical step in developing an effective tactile tile. It has been recommended that vision loss and visual functioning be classified by severity [39]. Table 2 and the Snellen test, which is typically performed by an ophthalmologist, classify visually impaired people into three levels of impairment: mild, moderate, and severe. Table 2. Classification of Visual Impairment by WHO Category

Displaying visual acuity at a distance Worse than:

Equal to or better than:

0 - The eyes are fine.

--

6/12 5/10 (0.5) 20/40

1 - minuscule problem with vision

6/12 5/10 (0.5) 20/40

6/18 3/10 (0.3) 20/70

2 - There is a moderate degree of vision loss.

6/18 3/10 (0.3) 20/70

6/60 1/10 (0.1) 20/200

170 Advances in Data Science-Driven Technologies

Prakash R. and Taduri

(Table 2) cont.....

Category

3 - Severe impairment of vision

Displaying visual acuity at a distance Worse than:

Equal to or better than:

6/60 1/10 (0.1) 20/200

3/60 1/20 (0.05) 20/400

Tactile Design To develop the adhesive tactile requirements, some criteria and inputs from visually impaired individuals are gathered, such as the most appealing color, foot sensitivity, and indentation profile. On the other hand, color blindness can occur in those who are visually impaired. The tactile color selected should contrast with the floor color [40]. As a result, individuals who are targeted are identified by their color contrast with the floor. After determining the color contrast, the foot sensitivity index is required. The ageing process will result in some foot problems. These individuals are blind to the tile's peaks and valleys. The foot is divided into three sections: metatarsal, Sesamoidal, and calcaneus. These areas are subjected to a sensitivity test by pressing them with the tip of the finger and recording the results. The following sections describe the stakeholder tests that were conducted to elicit critical feedback for the design of a tactile tile. Color Experimentation To ascertain which color is most appealing to all individuals, we conducted an experiment in which we projected the colors Red, Green, Blue, Yellow, and White onto the floor and asked elderly people to choose the most appealing colors, with the results being recorded. This is because these colors are among the most visually appealing in the visible spectrum. As detailed in the Experimental results section, data are gathered by asking elderly people questions and recording their responses on a Likert scale. Foot Sensitivity Test Individuals who walk on tactile indicators are compelled to do so, and individuals of varying ages will exhibit varying degrees of responsiveness to the various types of items they walk on. Certain individuals have a high sensitivity to their environment and surroundings, while others have a very low sensitivity. A sensitivity test was conducted on the feet of individuals of various ages by gently pressing on various areas of their feet with a finger at the Metatarsal, Sesamoidal, and Calcaneus segments of the legs and recording the results on a Likert scale. A 5mm bump height, they suggested, would be more acceptable to the target

People Mobility

Advances in Data Science-Driven Technologies 171

audience. Surface Texture Test It is necessary to determine the roughness value that is human-adaptive. To choose the tile's surface roughness so that it is both cognitively adaptive and antislip [40], a sensitivity test was conducted on older adults to determine the surface roughness in relation to the individual's foot sensitivity. Emery sheets with varying degrees of grit are used to evaluate people's responses. Such a texture can be applied to the tile if it is designed in such a way that it is instantly recognizable to anyone who walks on it. To determine their foot sensitivity, all study participants were given Emery sheets with three different GRIT levels (60, 80, and 120 grit), and their responses were recorded. Most respondents indicated that an 80-grit emery wheel was a viable option. Tactile Test Tiles are designed and developed in response to the feedback received from stakeholders. The created tiles are put to the test. The tiles are improved with the elderly's input and experience. Four/three tiles were developed in response to feedback and testing and were found to be the most popular with the target demographic. Even though there was a hazard design, people are very familiar with the existing warning tile [41]. The warning tile, straight tile, turn tile, and three-way junction tile are all illustrated in this diagram. While the design is suitable, there is one additional consideration: how far can people walk on the surface without it draining? It is critical to consider the materials that will be used to create the tiles in this corner to ensure that they are safe for people and are not easily destroyed. It was important to take all these factors into account when creating the tiles, which were printed using 3D printing materials like ABS, PLA, and TPU. The tiles are created using these resources, and feedback from stack holders is gathered. TPU-based tile is being considered for the design and development of tactile based on feedback. Based on feedback from stakeholders, the TPU-based tile is being evaluated for haptic design and development. We used Fusion Deposition Manufacturing (FDM), more commonly referred to as 3D printing, to create these TPU tiles. Due to the constant modification of the designs to achieve a satisfactory result, producing these tiles in a variety of designs using traditional manufacturing processes would be difficult. As a result, we chose 3D printing to create these tiles.

172 Advances in Data Science-Driven Technologies

Prakash R. and Taduri

(a)

(b)

(c)

(d)

Fig. (1). Different types of Tac Tiles used for Mobility (a) Warning Tile (b) Straight Tile (c) Turn Tile (d) 3way Junction tile

Tile Experiments The following stages describe the experiment. There were written and oral instructions in the elderly subjects' native language given to all participants, ensuring that everyone understood what was going on during the experiment. 1. Following an introduction of the experiment, volunteers were advised to familiarise themselves with the shape of each tile at the research site using their hands, feet, and white cane. Following that, the experiments were thoroughly described. Then there were practise walks and self-reporting exercises. When it appeared as though the job's intricacies were not fully mastered, the exercise was repeated; however, most individuals conducted each condition only once. 2. The usual trial was conducted according to the following protocols at random under each condition:

People Mobility

Advances in Data Science-Driven Technologies 173

a. The participant travelled to the trial's starting point under the watchful eye of an experimenter and awaited a verbal start cue. b. The experimenter, who stood around 1-2 metres behind the subject, shouted, “Start,” and the individual instantly began going straight toward the platform's edge. c. The participant came to a halt when they perceived themselves to be close to the platform edge because of their interaction with the tactile or the level floor following their interaction with the tactile. When no tiles were implanted, the individual proceeded until he was apprehended by the researcher. d. After coming to a halt, the individual maintained their position with their feet in the same position. During this period, the stopping distance was determined. e. Upon completion of the measurement, the subject walked to the location of the next experiment's start point under the supervision of the investigator. 3. Retrials were conducted when participants came to a halt in an area unrelated to the tiles, were unable to walk straight, or the experimenters were unable to accurately measure walking speed or step length. This complete technique is depicted in Fig. (2), which details the preceding operation.

Fig. (2). Tile Design and Test process

174 Advances in Data Science-Driven Technologies

Fig. (3). Mobility using Tactile by old age visual impaired person

Fig. (4). Mobility using Tactile by visual impaired person

Prakash R. and Taduri

People Mobility

Advances in Data Science-Driven Technologies 175

As you can see in the Figs. (3 and 4), experiments were performed on a wide range of blind and elderly individuals. The experiments' pseudocode/ algorithmic representation is as follows. Algorithm Mobility_Elderly_People Begin 1. Identify the target users. 2. Repeat the following steps 3 to 12 for all tiles. 3. Identify the most appealing color for elderly people. 4. Foot sensitivity test is conducted to identify foot-related problems. 5. Tile Surface texture test is conducted to know the roughness of the tile. 6. Tactile bumps information is explained to the target users in their native language. 7. Target users are asked to move on a tile to identify the texture. 8. Record the time taken to identify the tile. 9. If Tactile are identified, then. 10. Navigate from one end to another end. 11. Else. 12. Repeat steps 7 and 8 again in more detail way. 13. These experiments are represented on a Likert scale. End. RESULTS AND DISCUSSION A continuous monitoring instrument such as the Likert Scale was used from the start of each project to analyse its results and determine the next stages. The data from the stakeholders' Likert scales are utilised to determine the tile colour, the tile surface roughness, and the stakeholder's foot sensitivity. The test is carried out on elderly and blind people. According to their visual abilities, the

176 Advances in Data Science-Driven Technologies

Prakash R. and Taduri

elderly is divided into three groups. Users can be classified as mild, moderate, severe, or profound. Each participant was instructed to identify the indicator with their foot, and their responses were tallied using a Likert scale. They are then instructed to walk on these tiles to ensure their safety while navigating the indoor environment. With the assistance of (Fig. 5), we can modify questions 1 and 5. These questions concern the vibrant hues of red and yellow. We chose yellow for the tiles based on this information and current research on the colour that humans find most appealing (which is yellow) [39, 40].

Target User Responses in %

Color Test 70% 60% 50% 40% 30% 20% 10% 0% Strongly Agree

Agree

Neither Agree nor Disagree

Disagree

Likert 5 Point Scale Question 1 Question 2 Question 3 Question 4 Question 5 Question 6 Question 7 Question 8

Fig. (5). Result of Color Test

Fig. (6). Result of Surface Texture

Strongly Disagree

People Mobility

Advances in Data Science-Driven Technologies 177

For the sake of the elderly, avoid using a slippery tile. In accordance with (Fig. 6), the created and developed tile isn't slippery, so users can just walk on it. Foot problems are also brought on by ageing and the inability to feel bumps on the soles of the feet. As a result, a foot sensitivity test is carried out to find out whether the bare bottoms of the feet have difficulty penetrating at very low heel levels. According to the data in Fig. (7), many people do not have foot problems.

Fig. (7). Result of Foot Sensitivity Test

Fig. (8) shows the results of our stakeholder feedback after we designed and built three tile prototypes. As a result, the third prototype makes it simpler for users to spot patterns and move around the interface. Considering these findings, we believe visually impaired, and elderly people will have an easier time navigating their homes with the third prototype tactile than the other two have. After analyzing all the data, with the help of (Fig. 9), it was found that users took a wide range of times to identify warning tiles, with the average time taken being 17 seconds, and that for directional tiles taking 20, 35, 35 seconds. Stakeholders were asked to walk along a path that had been laid out following the identification of the issues. The duration of their journey was tracked, and the following results can be found in Fig. (10): Overall, the trip takes about 70 seconds on average.

178 Advances in Data Science-Driven Technologies

Fig. (8). Different types of tile prototypes and responses from stakeholders

Fig. (9). Tile Identification Time

Prakash R. and Taduri

People Mobility

Advances in Data Science-Driven Technologies 179

Fig. (10). Travel Time

CONCLUSION This article's goal is to provide elderly and visually impaired readers with a reliable navigation system for use in their own homes. They can use their foot to detect haptic signals on the device, which are self-explanatory. This sensor knowledge allows the user to move around inside a building. Home environments change frequently or only occasionally for people who live in leased residences (such as renters). The adhesive tactile can be easily adjusted for navigation by the aged and partially sighted. The proposed device is created and manufactured using ABS and TPU materials, as well as 3D printing. With the help of readily available materials and 3D printing technology, we developed surface indicators with a textured appearance. As 3D printing technology improves and materials become more readily available, low-cost multi-modal tactile surface indicators for the mobility of the elderly can be developed. CONSENT TO PUBLISH The authors affirm that human research participants provided informed consent for the publication of the images in Figs. (3 and 4).

180 Advances in Data Science-Driven Technologies

Prakash R. and Taduri

CONFLICT OF INTEREST The Department of Science and Technology, Govt. of India, funds this study (SEED/TIDE/035/2015). ACKNOWLEDGEMENTS The authors thank the reviewers for their constructive criticism and suggestions, as well as the Management, Principal and Staff of SR Engineering College in Warangal Urban for their ongoing support, which has helped to improve the quality of this paper. REFERENCES [1]

R.R.A. Bourne, S.R. Flaxman, and T. Braithwaite, Vision Loss Expert Group, "Magnitude, temporal trends, and projections of the global prevalence of blindness and distance and near vision impairment: a systematic review and meta-analysis", Lancet Glob. Health, vol. 5, no. 9, pp. e888-e897, 2017. [http://dx.doi.org/10.1016/S2214-109X(17)30293-0] [PMID: 28779882]

[2]

S. Swaminathan, and Y. Yim, "From Tactile to NavTile: Opportunities and Challenges with MultiModal Feedback for Guiding Surfaces during Non-Visual Navigation", In: Human Factors in Computing Systems (CHI ’21).Yokohama Japan, 2021, pp. 1-13.

[3]

R. Tapu, B. Mocanu, and T. Zaharia, "Wearable assistive devices for visually impaired: A state of the art survey", Pattern Recognit. Lett., vol. 137, pp. 37-52, 2020. [http://dx.doi.org/10.1016/j.patrec.2018.10.031]

[4]

S.R. Nyman, B. Dibb, C.R. Victor, and M.A. Gosney, "Emotional well-being and adjustment to vision loss in later life: a meta-synthesis of qualitative studies", Disabil. Rehabil., vol. 34, no. 12, pp. 971981, 2012. [http://dx.doi.org/10.3109/09638288.2011.626487] [PMID: 22066708]

[5]

J. Desrosiers, M.C. Wanet-Defalque, K. Témisjian, J. Gresset, M.F. Dubois, J. Renaud, C. Vincent, J. Rousseau, M. Carignan, and O. Overbury, "Participation in daily activities and social roles of older adults with visual impairment", Disabil. Rehabil., vol. 31, no. 15, pp. 1227-1234, 2009. [http://dx.doi.org/10.1080/09638280802532456] [PMID: 19802927]

[6]

N.A. Giudice, B.A. Guenther, T.M. Kaplan, S.M. Anderson, R.J. Knuesel, and J.F. Cioffi, "Use of an Indoor Navigation System by Sighted and Blind Travelers", ACM Trans. Access. Comput., vol. 13, no. 3, pp. 1-27, 2020. [http://dx.doi.org/10.1145/3407191]

[7]

L. de Witte, E. Steel, S. Gupta, V.D. Ramos, and U. Roentgen, "Assistive technology provision: towards an international framework for assuring availability and accessibility of affordable highquality assistive technology", Disabil. Rehabil. Assist. Technol., vol. 13, no. 5, pp. 467-472, 2018. [http://dx.doi.org/10.1080/17483107.2018.1470264] [PMID: 29741965]

[8]

W. Elmannai, and K. Elleithy, "Sensor-Based Assistive Devices for Visually-Impaired People: Current Status, Challenges, and Future Directions", Sensors (Basel), vol. 17, no. 3, pp. 565-580, 2017. [http://dx.doi.org/10.3390/s17030565] [PMID: 28287451]

[9]

S. Real, and Araujo, "Navigation Systems for the Blind and Visually Impaired: Past Work, Challenges, and Open Problems", Sensors (Basel), vol. 19, no. 15, pp. 3404-3419, 2019. [http://dx.doi.org/10.3390/s19153404]

[10]

N. Mizukami, K. Fujinami, H. Ohno, and H. Suzuki, "Research on utilization of tactile tiles and behavior of visually impaired persons on a railway platform", Quarterly Report of RTRI, vol. 43, no. 2,

People Mobility

Advances in Data Science-Driven Technologies 181

pp. 58-62, 2002. [http://dx.doi.org/10.2219/rtriqr.43.58] [11]

"Specifications for making buildings and facilities accessible to and usable by physically handicapped people", American National Standard, ANSI A117.1-1980, New York, 1980.

[12]

"Building’s facilities - Providing accessibility and usability for physically handicapped people", American National Standard, ANSI A117.1 – 1986, New York., 1986.

[13]

"Design for access and mobility – Part 4.1: Means to assist the orientation of people with vision impairment - Tactile ground surface indicators", Australian/New Zealand Standard, AS/NZS 1428-4, 2009.

[14]

Accessibility Guidelines for Buildings and Facilities, Americans with Disabilities Act. ADAAG, 1991.

[15]

"Accessible and usable buildings and facilities", American National Standard Institute, ICC, vol. ANSI117, p. 1, 1998.

[16]

A Europe accessible for all, European Commission., 2003.

[17]

"Assistive products for blind and vision impaired persons Tactile walking surface indicators", Japanese Industrial Standard., 2012.

[18]

A. Tennøy, "Aud Tennøy, Kjersti Visnes ksenholt, Nils Fearnley, and Bryan Matthews Standards for usable and safe environments for sight impaired, in conf", The Institution of Civil Engineers-Municipal Engineer., pp. 24-31, 2015.

[19]

L. Hakobyan, J. Lumsden, D. O’Sullivan, and H. Bartlett, "Mobile assistive technologies for the visually impaired", Surv. Ophthalmol., vol. 58, no. 6, pp. 513-528, 2013. [http://dx.doi.org/10.1016/j.survophthal.2012.10.004] [PMID: 24054999]

[20]

M.H.A. Wahab, A.A. Talib, H.A. Kadir, A. Johari, A. Noraziah, R.M. Sidek, and A.A. Mutalib, "Smart Cane: Assistive Cane for Visually impaired people", International Journal of Computer Science Issues, vol. 8, pp. 21-27, 2011.

[21]

A. Amedi, and S. Hanassy, Infra-Red Based Devices for Guiding Blind and Visually Impaired Persons, 2011.

[22]

S. Bharambe, R. Thakker, H. Patil, and K.M. Bhurchandi, "Substitute Eyes for Blind with Navigator Using Android", The India Educators Conference (TIIEC), pp. 38-43, 2013. [http://dx.doi.org/10.1109/TIIEC.2013.14]

[23]

Y. Yi, and L. Dong, "A design of blind-guide crutch based on multi-sensors", Fuzzy Systems and Knowledge Discovery (FSKD)., pp. 15-17, 2015.

[24]

S. Aymaz, and T. Çavdar, "Ultrasonic Assistive Headset for visually impaired people", conf. Telecommunications and Signal Processing (TSP)., pp. 27-29, 2016.

[25]

R. Agarwal, N. Ladha, M. Agarwal, K.K. Majee, A. Das, S.A. Kumar, S.K. Rai, A.K. Singh, S. Nayak, and S. Dey, "Low-cost ultrasonic smart glasses for blind", Electronics and Mobile Communication Conference (IEMCON), pp. 210-213, 2017.

[26]

D. De Alwis, and Y.C. Samarawickrama, "Low-Cost Ultrasonic Based Wide Detection Range Smart Walking Stick for Visually Impaired", In: On Multidisciplinary Approaches., 2016, pp. 123-130. 2016, pp. 123–130.

[27]

M. Elgendy, T. Guzsvinecz, and C. Sik-Lanyi, "Identification of Markers in Challenging Conditions for People with Visual Impairment Using Convolutional Neural Network", Appl. Sci. (Basel), vol. 9, no. 23, pp. 5110-5127, 2019. [http://dx.doi.org/10.3390/app9235110]

[28]

A.L. Petsiuk, and J.M. Pearce, "Low-Cost Open Source Ultrasound-Sensing Based Navigational Support for the Visually Impaired", Sensors (Basel), vol. 19, no. 17, pp. 3783-3803, 2019. [http://dx.doi.org/10.3390/s19173783] [PMID: 31480451]

182 Advances in Data Science-Driven Technologies

Prakash R. and Taduri

[29]

L. Dunai, B.D. Garcia, I. Lengua, and G. Peris-Fajarnés, "3D CMOS sensor based acoustic object detection and navigation system for blind people. ", IEEE Industrial Electronics Society (IECON 2012)., pp. 25-28, 2012. [http://dx.doi.org/10.1109/IECON.2012.6389214]

[30]

A. Landa-Hernández, and E. Bayro-Corrochano, "Cognitive guidance system for the blind", IEEE World Automation Congress (WAC), pp. 24-28, 2012.

[31]

J.M. Benjamin Jr, "The laser cane", Bull. Prosthet. Res., vol. 10, pp. 443-450, 1974. [PMID: 4462934]

[32]

M.R.U. Saputra, and P.I. Santosa, "Obstacle Avoidance for Visually Impaired Using Auto-Adaptive Thresholding on Kinect’s Depth Image", Scalable Computing and Communications and Its Associated Workshops (UTC-ATC-ScalCom)., pp. 9-12, 2014. [http://dx.doi.org/10.1109/UIC-ATC-ScalCom.2014.108]

[33]

I. Ahlmark, D. Hakan Fredriksson, and K. Hyyppa, "Obstacle avoidance using haptics and a laser rangefinder", IEEE Workshop on Advanced Robotics and its Social Impacts (ARSO), 2013pp. 7-9

[34]

R. Shilkrot, J. Huber, C. Liu, P. Maes, and S.C. Nanayakkara, "Fingerreader: A wearable device to support text reading on the go", Human Factors in Computing Systems., pp. 2359-2364, 2014. [http://dx.doi.org/10.1145/2559206.2581220]

[35]

G. Olmschenk, C. Yang, Z. Zhu, H. Tong, and W.H. Seiple, Mobile crowd assisted navigation for the visually impaired [http://dx.doi.org/10.1109/UIC-ATC-ScalCom-CBDCom-IoP.2015.69]

[36]

B. Mocanu, R. Tapu, and T. Zaharia, "When Ultrasonic Sensors and Computer Vision Join Forces for Efficient Obstacle Detection and Recognition", Sensors (Basel), vol. 16, no. 11, pp. 1807-1822, 2016. [http://dx.doi.org/10.3390/s16111807] [PMID: 27801834]

[37]

A. Martinez-Sala, F. Losilla, J. Sánchez-Aarnoutse, and J. García-Haro, "Design, implementation and evaluation of an indoor navigation system for visually-impaired people", Sensors (Basel), vol. 15, no. 12, pp. 32168-32187, 2015. [http://dx.doi.org/10.3390/s151229912] [PMID: 26703610]

[38]

A. Pembuain, S. Priyanto, and L.B. Suparma, "The evaluation of tactile ground surface indicator condition and effectiveness on the sidewalk in Yogyakarta City, Indonesia", IATSS Res., vol. 44, no. 1, pp. 1-7, 2020. [http://dx.doi.org/10.1016/j.iatssr.2019.04.002]

[39]

"International Classification of Functioning, Disability and Health (ICF) Version 04/2019", World Health Organization., 2019.

[40]

Y. Kobayashi, R. Osaka, T. Hara, and H. Fujimoto, "How accurately people can discriminate the differences of floor materials with various elasticities", IEEE Trans. Neural Syst. Rehabil. Eng., vol. 16, no. 1, pp. 99-105, 2008. [http://dx.doi.org/10.1109/TNSRE.2007.910283] [PMID: 18303811]

[41]

K. Fujinami, N. Mizukami, H. Ohno, H. Suzuki, A. Shinomiya, O. Sueda, and M. Tauchi, "Tactile Ground Surface Indicator Widening and its Effect on Users’ Detection Abilities", Quarterly Report of RTRI, vol. 46, no. 1, pp. 40-45, 2005. [http://dx.doi.org/10.2219/rtriqr.46.40]

Advances in Data Science-Driven Technologies, 2023, 183-210

183

CHAPTER 9

Assistive Technology Trends, Challenges and Future Directions Nancy Jasmine Goldena1,* and Thangapriya1 Department of Computer Applications and Research Centre, Sarah Tucker College (Autonomous), Tirunelveli, Tamilnadu, India 1

Abstract: People with impairments frequently struggle to carry out daily activities alone or even with assistance. They encounter obstacles in the environment, movement, interaction, access to writings, personal health maintenance, handling medical issues and behavioral equality. One of the subjects that have received a lot of attention from researchers is computer-based Assistive Technology (AT). Disabled people utilize AT to tackle things previously practically impossible for them. Various forms of disabilities necessitate the use of AT, which can help people with disability to do their regular work. Therefore, these technological innovations have the power to play a substantial role in supporting huge segments of society to operate and lead a normal life. The fundamental goal of AT is to continually increase a person's ability to perform independently, hence improving their overall health. Individuals who use technological aids can lead healthy, dignified, independent and respectable lifestyles. On the whole, AT aims to enable disabled individuals to join nearly every facet of life, including at home, education and community, as well as to increase their opportunities for social interactions and meaningful employment. AT devices simply gives disabled individuals more freedom and control. The significance of AT and AT devices, current trends, approaches, limitations and some of the major challenges identified in previous assessments as well as recent research findings in the field of AT, are all effectively discussed in this chapter.

Keywords: Artificial intelligence, Assisted living, Assistive technology, Braille, Cognitive impairment, Daily living, Disability, Elderly care, Healthcare, Hearing impairment, Human activity recognition, Internet of things, Learning, Mobility challenges, Physically-challenged, Prosthesis, Research gaps for AT, Robotics, Sip and puff, Vision impairment.

* Corresponding author Nancy Jasmine Goldena: Department of Computer Applications and Research Centre, Sarah Tucker College (Autonomous), Tirunelveli, Tamilnadu, India; Tel: +91-9487081610; E-mail: [email protected]

Manoj Kumar M V, Immanuel Azaad Moonesar R.D., Ananth Rao, Pradeep N, Annappa, Sandeep Kautish and Vijayakumar Varadarajan (Eds.) All rights reserved-© 2023 Bentham Science Publishers

184 Advances in Data Science-Driven Technologies

Goldena and Thangapriya

WHERE ARE WE NOW WITH ASSISTIVE TECHNOLOGY? AT refers to assistive, adaptive and rehabilitative equipment for disabled and elderly individuals [1]. AT encourages greater autonomy by improving the AT device to complete activities that disabled people previously could not complete. The Evolution of AT The usage of AT devices and innovations has evolved over the years, as shown in Fig. (1). There have been key incidents in different historical periods that expanded the use of AT [2].

Fig. (1). Evolution of AT

Foundation Period (1800 – 1900) The period preceding 1900 is considered the foundation period. During this time, people were capable of surviving with physical impairments. Education systems for blindness, dumb and some other impairments existed throughout the early 1900s [3]. AT inventions made during the foundation period are listed in Table 1. Table 1. Inventions during the foundation period Year

Foundation Period Inventions

1829

Braille is a haptic communication system that consists of six raised dots and 64 combinations that can be read by visually impaired persons. It was officially published in 1929 and is still in use today.

1836

Taylor invented the first practical math instrument that could be utilized by people who were blind.

1869

The besica model for a manual wheelchair was originally patented and used in the United States during the War.

1877

The phonograph was invented by Thomas Edison.

1876

Alexander Graham Bell's invention is used to create the first wearable cochlear implant.

1892

The Braille keypad was developed by Frank Hall.

1898 Akouphone is the very first electric implant and compact enough to put in a pocket, are developed.

Establishment Period (1900-1972) Impairment disciplines were developed during this establishment period [3]. People’s perceptions of disabled people had shifted in a good direction. As many

Assistive Technology

Advances in Data Science-Driven Technologies 185

people were suffering traumas, the number of disabled people grew. Some of the AT inventions made during the establishment period are listed in Table 2. Table 2. Inventions during the establishment period Year

Establishment Period Inventions

1932

Harry Jennings created the first steel frame foldable wheelchair.

1935

The phonograph is deployed for both entertaining and learning in the form of talking books.

1936

The Voder is the first electronic voice synthesizer. It has a keypad and foot controls for controlling the machine and delivering sound.

1947

The hoover cane was created for soldiers who had been blind during the war.

1951

The Perkins Brailler typewriter was created to enable Braille typing. Writing Braille used to be a challenging task.

1952

Tellatouch, a deaf and blind communicating gadget, was introduced.

1960

To reduce their inferior complication, the first Paralympic Games were hosted in Rome. Sip and Puff technology, research by the University of Chicago, uses air pressure to regulate the device, similar to puffing through a straw. Paraplegics are the primary users of Sip and Puff systems nowadays.

1966

The Lasercane was invented, which produced light beams to detect and recognize, preventing clear movement.

1971

Optacon was promoted as a tool that would help blind individuals to read text.

Empowerment Period (1972-2010) Individuals with disabilities were granted the right to achieve their life goals during the empowerment period [3]. Many legislations have been passed to improve the rights of people with disabilities. During this time, many AT devices were developed to help people with disabilities gain independence and achieve their goals. During this period of empowerment, disabled people understood their “WILL TO SUCCEED”. AT inventions made during the empowerment period are listed in Table 3. Table 3. Inventions during the empowerment period Year

Empowerment Period Inventions

1976

The first computer software to detect printed letters was the Kurzweil Reading Machine. The first wearable voice generator was also developed in the same year.

1983

The augmented communication company is established and researchers chose to invent a technology, DynaVox, which allowed individuals to speak using only their eyes.

1992

Text-to-speech computer techniques are featured to assist people with disabilities in accessing printed texts.

186 Advances in Data Science-Driven Technologies

Goldena and Thangapriya

(Table 3) cont.....

Year 1996

Empowerment Period Inventions In education, students with hearing problems as well as other students in the class, benefit from FM amplification technology.

2009

SMART board makers have created a SMART table for students with motor impairments.

2010

Mobile applications such as voice recording are available on apple products such as the iPhone, iPod and iPad.

Technologically Sophisticated Period(2011-present) During this time, advanced AT devices were being produced. The research and manufacturing sectors developed new types of AT equipment that people with disabilities successfully used. AT devices and their technological advancements are listed in Table 4. In terms of technological advancement, AT devices are modified. Two specific areas of AT are Assistive Technology Services (ATS) and Assistive Technology Devices (ATD). Table 4. Technological advancement of AT devices ATD

Technological Advancement

Wheelchairs

Newer advancements in wheelchair design allow wheelchairs to climb stairs, travel off-road or use other add-ons such as hand bikes or power assistance.

Walkers

Bipedal nanowalkers can walk in both forward and backward directions on a given track.

Prosthesis

Artificial heart valves are widely used inside the body, while artificial hearts and lungs are less popular but still under development.

Exoskeleton

The robotic exoskeleton provides support for the arm, hip and thighs as well as assists movement for grasping and gripping heavy objects while reducing back strain.

Screen reader

Visually impaired people utilize screen magnifiers, screen readers, desktop video magnifiers, braille embossers and other advanced software [6].

Personal Emergency Response System

Personal Emergency Response Systems (PERS) are a kind of AT that includes electronic sensors linked to an alerting system to help caretakers in managing danger and allow sensitive persons to stay home uninterrupted [6].

Assistive Listening devices

By removing unnecessary background noises and distractions, people with hearing impairments can focus on a speaker or subject, making auditoriums, classrooms and meetings much more accessible.

Educational software

Educational software is designed to help people who struggle with reading, learning, understanding and planning.

Car for blind

A car for the blind is currently being created by engineer Dennis Hong who will be able to monitor and observe its environment while giving sound warnings, pulsating gloves and seating vibrations [2].

Driverless Car

Google's self-driving cars, which rely on Artificial Intelligence and Google Street View, are currently in development for people with cognitive disabilities, providing them with a safe and independent mode of transportation [2].

Assistive Technology

Advances in Data Science-Driven Technologies 187

(Table 4) cont.....

ATD

Technological Advancement

ND Assistive

ND Assistive opened the Home First Demonstration Center in Fargo, North Dakota, a simulated home equipped with AT.

Alexa

Alexa is Amazon's voice-based AI-powered digital assistant that enhances the entire ecosystem of smart devices. The Echo device uses speech recognition to carry out the tasks or commands given by the user.

Seeing AI

The Google Home app aids in the setup and control of Google Nest or Home speakers and displays, as well as Chromecast. A single app can control thousands of compatible lights, cameras, speakers and other devices as well as check reminders and recent notifications.

Voice Control

Seeing AI, the first free software on Apple's platform released by Microsoft, uses the device camera to identify people and things and speaks out those things for the visually impaired.

Live Transcribe

Google has released Voice Control for Android smartphones.

Any service that supports a person with a disability in selecting, acquiring or using AT equipment is referred to as an Assistive Technology Service and any item or piece of equipment used to strengthen, maintain or develop the abilities of a person with a disability, whether purchased commercially or customized, is termed as Assistive Technology Device [3 - 5]. Legal Mandates Many legal regulations have been enacted to assist people with disabilities who are in need. Later, double-blind evaluations and strict legislative rules governing people's lifestyles, employment and other necessities were put in place. Many of them are unaware of the government's policies regarding people with disabilities. As a result, everyone should understand the fundamental law of disability management. Some of the legislations for the disabled are in Table 5 [7]. Table 5. Essential legislations of AT Legislations

Description of legislations

Individuals with Impairments Education Act (IDEA)

The Education for All Handicapped Children Act (between 1975 and 1990) was reauthorized, and its name was changed to IDEA. IDEA defines AT services as “any service that significantly assists a disabled person in the purchase, planning, fitting, personalizing, adaptation, deploying, preserving and correcting” of AT.

The AT Act

The AT Act (P.L. 105-394) was passed by Congress in 1998 to provide excellent opportunity to, affordability of and financing for AT for all people with disabilities, including several toddlers, and it was revised in 2004.

188 Advances in Data Science-Driven Technologies

Goldena and Thangapriya

(Table 5) cont.....

Legislations

Description of legislations

Every Student Succeeds Act (ESSA)

Every Student Succeeds Act (ESSA) is an US law that covers K–12 public education policy in December 2015. ESSA supports the appropriate use of assistive and educational technology to teaching-learning process from The Center on Technology and Impairment.

Impairment Rights Ohio

Impairment Rights Ohio (IRO) is a non-profit group whose aim is to advocate for individuals with disabilities in Ohio's economic, social and rights under the law. This mainly includes supporting the juvenile and criminal judicial systems with issues such as abuse, neglect, discrimination, access to AT, special education, housing, employment, community integration, voting and rights protection.

The Americans with Disabilities Act

The Americans with Disabilities Act (ADA) of 1990 is a civil rights legislation that prevents discrimination against people with disabilities in employment, government and local, places of public accommodation, business centers, transportation and telecommunication services.

Fair Housing Act

On September 13, 1988, the Fair Housing Amendments Act (FHAA) was passed into law, and it went into effect on March 12, 1989. It is illegal to refuse housing to a renter or buyer because of a physical or mental disability. In order to accommodate people with disabilities, owners must offer reasonable adjustments to their policies. If necessary, tenants are also permitted to make reasonable access-related alterations to the property.

The Television Decoder Circuitry Act

The Communications Act of 1934 was revised by the Federal Communications Commission (FCC) in 1990. The law took effect on July 1, 1993, and says that all television receivers with visual screens of 13 inches or larger must comply.

Telecommunications Act

The Telecommunications Act of 1996 is the first substantial revision of telecommunications legislation in about 62 years. This act has the potential to alter our working, living and learning habits. Persons with disabilities must be able to access and use telecommunications equipment and services under this law.

Workforce Investment Act

Workforce investment act suggests to implement technology and its application in job planning as well as the acquisition and retention of individuals with disabilities in the vocational rehabilitation process.

Section 508 of the Rehabilitation Act

All technological and method of determining established and used by any Federal government agency must be accessible to people with disabilities, according to Section 508 of the Rehabilitation Act. Websites, video and audio recordings, electronic books, television shows and other forms of media fall under this category.

IMPORTANCE OF ASSISTIVE TECHNOLOGY People with a range of disabilities are the primary users of AT, thus, the disability must first be identified. The categories of disability are cognitive, motor, visual and auditory, where motor, visual and auditory fall under physical disability. The categories of disability are shown in Fig. (2), and their symptoms are shown in Fig. (3).

Assistive Technology

Advances in Data Science-Driven Technologies 189

Fig. (2). Types of Disability

Fig. (3). Categories and Symptoms of Disabilities

Mental dysfunctions, dysfunctions that lead people to learn differently than people

190 Advances in Data Science-Driven Technologies

Goldena and Thangapriya

without disabilities or illnesses that interfere with a person's thinking process are all examples of cognitive disability [8]. The whole or partial loss of body parts, the presence of organisms that cause illness in the body, and abnormalities and disorders of body parts are all examples of physical disability [9]. Cognitive Disability When a person struggles to remember things, learn new things, focus or make important decisions in daily life, it is called cognitive disability. People with moderate cognitive disabilities may perceive changes in their cognitive skills but continue to be able to carry out their daily activities. Severe impairment can result in the loss of the ability to understand the sense as well as the ability to speak or write, making it impossible to live independently. Computers and AT are becoming more prevalent in the teaching of students with special needs. Cognitive Assistive Technology (CAT) can help students overcome some of the functional limitations posed by disability, allowing them to read, write and communicate more effectively. In terms of education productivity, the availability of AT can help to “FAIRNESS AND EQUALITY”. These technologies, together with the growing use of computers in business, have opened up new independent living and career prospects for pupils as well as improved their perceptions of their own future development and role in life [1]. Using specialized software, individuals with speech difficulties may communicate, and students with learning disabilities can read and write. AT can help with many aspects of life for an elderly adult, such as daily household duties, individual safety and well-being, memorizing appointments, keeping in contact with family and friends and reducing stress on caretakers, improving their quality of life and staying at home longer [6]. Various types of cognitive disabilities are given in Table 6. Table 6. Symptoms of Cognitive Disability Symptom

Description

Attention

People's attention does not permit them to concentrate on information in order to develop memories.

Perception

Disability to capture, process and actively make meaning of the information received by senses.

Memory

Unable to acquire, store, retain and retrieve the information later.

Orientation

Mental dysfunction involves an unawareness of time, location and person in three dimensions.

Dementia

Amnesia or memory problems, loss of orientation, confusion and difficulties with home hygiene.

Assistive Technology

Advances in Data Science-Driven Technologies 191

(Table 6) cont.....

Symptom

Description

Learning

Learning disorders are caused by genetic and neurobiological factors that impact one or more cognitive processes associated with learning via altering brain functioning.

Knowledge Representation

Unable to make context clues from knowledge easier.

Cerebral Vascular Incidents

Signs of vascular cognitive impairment normally arise gradually and worsen as blood vessel damage increases. Symptoms might range from minor memory, focus and thinking issues to more serious and broad issues.

Developmental

The state of a person whose intellectual capacity and adaptive behavior are significantly below the expected average for their age.

Recognition

A major recognition difficulty that almost certainly makes it impossible to carry out a significant portion of everyday duties.

Daily organization

Difficulty performing daily tasks such as cooking, eating, studying, working, and maintaining personal health.

Education

Children with learning disabilities have poor social development, including inferior and negative cognitive development, low learning levels, low selfesteem, despair, fear, aggressive and avoidance behavior in moods.

Motor Disability A physical disability is defined as a loss or limited function in muscular control, movement or mobility, as well as an injury or infirmity that prevents normal physical functioning. AT promotes the independence and autonomy of both individuals and those around them by reducing dependency and supporting an independent life. Specially designed AT devices provide the tools necessary for a disabled person to succeed when their ability to execute is limited. Manual wheelchairs improve educational and job opportunities while lowering healthcare expenditures by reducing the risk of pressure sores and contractures [6]. Various symptoms of motor disability are given in Table 7. Table 7. Symptoms of Motor Disability Symptom

Description

Spinal cord injury

Serious spinal injuries can reduce or eliminate control and motion in various portions of the body. People with severe spinal injuries are unable to walk because they can no longer control their legs as well as they once could.

Lost / Damaged limbs

Any damage to a limb, either a leg or arm or to the toes and fingers. Limb injuries include broken bones, dislocations, sprains and strains.

Cerebral palsy

Cerebral palsy produces spastic or flaccid muscles, weak reflexes, poor coordination, involuntary motions, poor posture and poor balance.

Muscular Dystrophy

Loss of muscle strength can lead to spasms and cramping. When the chest becomes weak, the spine curves and mobility begins to deteriorate.

192 Advances in Data Science-Driven Technologies

Goldena and Thangapriya

(Table 7) cont.....

Symptom

Description

Physically disabled

Major motor disorders, such as walking and impaired fine motor skills, such as manipulating objects by hand.

Parkinson’s disease

Movement illness that affects the brain such as tremor, slowdowns of movement, tight muscles, shaky walking and difficulty with balance and coordination.

Essential tremor

Uncontrollable and rhythmic shaking condition of the nerve system.

Visual Disability Visual disability is defined as vision loss that results in problems that cannot be corrected with common methods such as eyeglasses and medications. Illness, trauma or a congenital or degenerative disorder can all cause visual disability. AT helps people read using Braille, text-magnifying devices, voice output and also the computer uses voice recognition software to respond to voice instructions [6]. Various symptoms of visual disability are given in Table 8. Table 8. Symptoms of Visual Disability Symptom

Description

Low or No vision

Age-related macular degeneration, diabetes and glaucoma are all prevalent causes of vision disability. Low vision can also be caused by eye cancer, albinism, brain injury or genetic eye problems, including retinitis pigmentosa.

Blindness

Loss of vision that cannot be rectified with glasses or contact lenses.

Generalized haze

The illusion of a film or glare that may spread across the entire seeing field.

Extreme light sensitivity

Standard amounts of lighting exceed the visual system, resulting in a washed-out vision and glare impairment.

Night blindness

Unable to see clearly at night or in dim light.

Auditory Disability Auditory disability is a brain disorder that affects the ability to hear sounds and understand spoken language. AT used by the auditory disabled can help improve language skills. Without auditory AT devices, people with hearing loss have significantly limited, especially in education and employment opportunities [6]. Various symptoms of auditory disability are given in Table 9. Table 9. Symptoms of Auditory Disability Symptom

Description

Aging

The gradual loss of hearing in both ears is a result of aging.

Injury

Damage to the eardrum or ossicular chain, disruptions in intralabyrinthine fluid and cochlea as a result of a head injury.

Assistive Technology

Advances in Data Science-Driven Technologies 193

(Table 9) cont.....

Symptom

Description

Excessive noise exposure

Hearing loss due to the eardrum vibration to the point of fracture or damage as a result of the high sound pressure.

Viral infectious

Hearing loss due to viral infections, particularly cytomegalovirus.

Shingles

Ramsay hunt syndrome occurs when a shingles outbreak hits the facial nerve near one of the ears, causing facial paralysis and hearing loss in the affected ear in addition to the painful shingles rash.

Acoustic tumors

Causing a dangerous build-up of fluid in the brain.

Heredity

Due to genetic reasons, genes predispose to hearing loss as aging or as a result of noise, medicines or infections.

Perforation

Hearing loss due to a ruptured eardrum. Infection due to perforation of the middle ear.

Abnormal growth

Cholesteatoma, a cyst-like growth that forms in the center of the eardrum. It can be congenital, and caused by a chronic ear infection.

Pus build-up

An ear infection in the ear canal or middle ear.

APPROACHES AND CRITICISMS IN THE CURRENT STUDY OF ASSISTIVE TECHNOLOGY Approaches of AT AT approaches fall into three categories based on their technical sophistication such as low-tech, mid-tech and high-tech, as shown in Fig. (4). Low-tech AT devices refer to electronics that do not require much training, are often less expensive and lack complicated. These devices have limitations in terms of the amount of data that can be saved and how that data is displayed to the user Midtech AT devices need training and need to be affordable. High-tech AT devices are completely electronic and require extensive training before use [3].

Fig. (4). AT Approaches

194 Advances in Data Science-Driven Technologies

Goldena and Thangapriya

AT for Cognitive Disability AT for cognitive disability improves a person's educational prospects, social interactions and possibility for meaningful work. It also encourages students to participate in learning activities in the least restrictive setting possible. AT is a tool to help students benefit from the general education curriculum and access extracurricular activities in schools, homes and workplaces. Teachers can support students to express themselves and speak by using assistive technology tools for developing effective communication skills by encouraging positive thinking and student participation. Speech recognition software and a speech-producing device are two of the most common AT devices used for communication today. Instructors can use Speech Recognition Software to talk to the computer through a microphone, and the uttered words will appear as text [3]. Disabled people can benefit from this type of AT device because it allows them to select words from a monitor that were not detected when speaking. It is beneficial to disabled people who have difficulty with motor skills, mobility or oral language skills. Reading aloud with dual color highlighting, spoken dictionary, picture dictionary, translator and fact finder, study skills highlighters and collect highlights, vocabulary list builder, annotations and navigational tools are all included in Google Read and Write. Therefore, it encourages disabled people to read, write and express themselves more confidently. There are numerous AT devices available today that can assist students in learning unique skills. Students with writing, listening, reading, organizational and memory problems can use spell checkers, tape recorders, proofreaders, optional character recognition, electronic math worksheets, variable speech control, FM hearing systems, conversation calculators, personal data managers and so on [3]. They can also benefit from tools such as freeform databases and prewriting organizers. These AT devices boost the confidence of struggling readers and those with learning disabilities. A list of AT devices that can be used to treat a variety of cognitive disability is given in Table 10. Table 10. AT devices for Cognitive Disability Devices Approach

Description

High lighter Low-tech

Helps disabled people to pay attention more actively.

Clock Low-tech

Specially built for disabled people to keep track of time.

Calendar Low-tech

To alert disabled people to complete a task.

Assistive Technology

Advances in Data Science-Driven Technologies 195

(Table 10) cont.....

Devices Approach

Description

Personal digital assistant Low-tech

A compact, mobile, handheld device that has computer and information storing and retrieval capabilities for personal and is frequently used to manage schedules, calendars and address book information on hand.

Tape recorders Low-tech

Used to record television shows for live streaming for disabled people.

Pagers Low-tech

A wireless telecommunications device that receives and displays alphanumeric or voice messages. It also is known as a beeper, bleeper or pocket bell.

Talking calculator Mid-tech

Helps people with dyscalculia. Checking assignments, reading numbers and performing math are all made easier with the device.

Tactile geometric kit Mid-tech

A set of thermoformed raised-line drawings representing geometric shapes, forms and interconnections for the visually disabled.

Speech recognition software Mid-tech

Computer software to convert human speech into text.

Google read & write High-tech

Used as a speech recognition system.

Electric tablets High-tech

Disabled people can use tablets for calendars, voice-to-text and a variety of other tasks.

AT for Motor Disability Mobility aids allow people to move around their surroundings by modifying vehicles for travel. For operating a computer, people with limited hand functions can use a keyboard with large keys or a special mouse. People with mobility issues use high-tech equipment to do work independently, which allows them to manage and input data using a handle that they can easily move with their mouths. By allowing people to do tasks independently using AT tools, mobility tools and devices can easily be integrated into the educational curriculum [3]. A list of AT devices that can be used to assist various motor disabilities is listed in Table 11. Table 11. AT devices for Motor Disability Devices Approach

Description

Canes Low-tech

Used to alleviate pain and enhance balance

Adapted pencil Low-tech

Helps to alter the diameter, shape and texture of writing tools for people with fine motor control in writing more comfortably and comprehensibly.

Pencil grip Low-tech

Helps to attain better hand posture and flexibility when writing and leads to improved handwriting.

196 Advances in Data Science-Driven Technologies

Goldena and Thangapriya

(Table 11) cont.....

Devices Approach

Description

Adapted eraser Low-tech

Built specifically for the purpose of erasing disabled individuals.

Adapted books Mid-tech

Any book that has been adjusted in some way to make it more accessible to a person who has trouble with classical literature.

Raised line graph Mid-tech

Used to assist struggling hand writers in staying within the writing lines.

Walkers Low-tech

Designed to assist in walking with balance and strength while standing.

Crutches Low-tech

Used to keep a person upright and can be worn by people who have short-term or long-term disabilities.

Electrical wheelchair Low-tech

Used by a person who cannot walk.

Prosthetics Low-tech

Used for neural damage, amputation and other mobility-related issues.

Orthotics Low-tech

Designed to provide a stable foundation for training, increase gait efficiency and correct or prevent deformity.

Slant board Low-tech

Beneficial for relief from pain from tightness caused by poor running mechanics and muscle imbalance.

Scooter Mid-tech

Designed for more convenience to use and operate.

Adapted switches Mid-tech

Mechanism that enables persons with mobility limitations to use technology and operate electrical appliances.

Adapted seating Mid-tech

Improves sitting posture and postural control in mobility-impaired people.

AT for Visual Disability Orientation and mobility are critical skills for every day and independent functioning in people who have lost their vision. People with visual disabilities and blindness encounter numerous obstacles. There are a number of ATDs that can help with writing tasks. For persons who are blind or visually challenged, many modern electrical and computational technologies can be quite useful. Specifically, those who are visually impaired, partially sighted or blind have a difficulty to track and pay visual attention to objects. AT strengthens visual tracking skills to make learning and day-to-day life easier for these learners. Most visual tracking tools include a sliding feature, and some utilize lights and an auditory component to keep the disabled engaged. Blind people can use software that reads text on the screen in a computergenerated voice, and people with low vision can use software that enlarges the

Assistive Technology

Advances in Data Science-Driven Technologies 197

screen content [1]. Mathematical assistive technology is especially useful for visually impaired students to easily learn math concepts. Text and audio aids such as Braille textbooks and calculators can help improve student access to math textbooks. Nevertheless, tactile aids and tactile techniques can help promote concrete mathematical understanding. In low-contrast situations, a lengthy symbol can be used to detect kerbs, entrances or obstructions. The symbol cane is generally used to alert the general public that the person has low vision or has a visual impairment [1]. A list of AT devices that can be used to assist various visual disability is listed in Table 12. Table 12. AT devices for Visual Disability Devices Approach

Description

Adaptive paper Low-tech

Can increase legibility, writing size and keep writing on the baseline.

Braille compass Mid-tech

Designed to assist visually impaired people in finding directions.

Braille ruler Mid-tech

A low-cost measuring ruler for the visually disabled.

Braille protractor Mid-tech

A low-cost measuring protractor for the visually disabled.

Braile cube Mid-tech

A classic toy to the needs of blind people, with the haptic experience allowing them to feel the colors.

Routing tools High-tech

The majority of wheelchairs are electric and include an innovative routing tool that makes life easier for impaired persons.

Mouth stick High-tech

An assisted device for those having cerebral palsy as well as those who are unable to move their hands due to quadriplegia.

Tracking device High-tech

A mobility device that utilizes space technology to assist in leading the direction.

AT for Auditory Disability AT communication aids can help people who have difficulty speaking or hearing. Many new electrical and computational technologies are effective tools for those who are deaf or hard of hearing. People with auditory disability can use a TTY (text telephone), a special device that allows hearing-impaired or speech-impaired people to communicate using the telephone by typing messages back and forth instead of speaking and listening. A hearing aid is another small electronic device that is worn behind the ear or in the ear canal. It amplifies sounds so the disabled

198 Advances in Data Science-Driven Technologies

Goldena and Thangapriya

person can interact and participate in everyday activities more easily. There are three pieces to a hearing aid. The noises are picked up by a microphone, which turns them into electric impulses, then transmitted to an amplifier. The amplifier boosts the signal's power and sends it through a speaker into the ear. A list of AT devices that can be used to assist various auditory disability is listed in Table 13. Table 13. AT Devices for Auditory Disability Devices Approach

Description

Text Telephones Low/Mid-tech

Helps hearing or speech-impaired to communicate over the phone by allowing them to write messages back and forth instead of talking and listening.

FM systems Low/Mid-tech

A unique wireless device that improves hearing in noisy environments.

Infrared systems Mid-tech

An audio technology that aids in hearing and communication for those with hearing disability.

Alerting devices High-tech

Connected to a doorbell, phone, alarm, blinking light to signal somebody with hearing disability that something is happening.

Assistive listening devices High-tech

Used in combination with a hearing aid or even an auditory implant to improve a user's ability to listen selected sounds when there is a lot of background noise.

Alternative communication A device that aids communication for people who have a speech or language devices disability. High-tech

Criticisms in Implementing AT AT is not a magic cure; if too much attention is placed on technology rather than on how people with disabilities will react, it is a failure. Failure is vital to avoid because it is a setback for people using AT devices and can lead to unfulfilled expectations and disappointment. It is indeed crucial to prevent this as it frequently leads to a waste of money and time. This problem occurs when too much importance is put on technology as a solution, with little consideration being given to how the technology will function in a busy atmosphere or how the disabled will react to it [6]. The usage of AT may be restricted due to the following factors: • Limited number of solutions available to assist the disabled in realizing their full potential. • Some high-tech AT elements require a lot of learning that may exceed the cognitive ability or physical endurance of a person with a disability. • People with disabilities may not “just buy” AT if it shows their disability.

Assistive Technology

Advances in Data Science-Driven Technologies 199

• Electronic communication technology may not allow people with disabilities to participate in the normal process of communication. • Professionals don't always know better. • The instinctive knowledge of caregivers of people with disabilities is sometimes valuable, but it carries less weight than the opinions of “experts.” • There is no such thing as a “final solution.” • The circumstances and requirements of disabled persons are always changing. • New items enter the market, and technological advancements may need regular re-evaluation. • Some AT is quite costly. • A considerable number of people with disabilities are left without access to assistive technology. • AT deployment in rural areas is greatly hampered compared to urban areas. Some countries have recognized the need for AT support for people with disabilities through policies, but they have yet to develop a clear provision strategy to aid in the acquisition of assistive products. • Difficult to find a technician who can fix ATD, particularly wheelchairs, in rural regions. LIMITATIONS AND CHALLENGES IN ASSISTIVE TECHNOLOGY Many advances have been made, but we still have a long way to go in terms of adopting AT. There are numerous restrictions in areas such as awareness, governance, services, products, environments, human resources and finance, as shown in Fig. (5). Even in wealthy countries, implementing AT is difficult due to the wide range of disabilities [7].

Fig. (5). Limitations of AT

200 Advances in Data Science-Driven Technologies

Goldena and Thangapriya

Lack of Awareness Many people with disabilities and their families are unaware of the many products and services available to them. This makes it difficult for disabled people and their families to understand which assistive technologies are available and acceptable, as well as how they may assist. Lack of Governance According to a survey “Global survey on government action on the implementation of the Standard Rules on the Equalization of Opportunities for Persons with Disabilities,” Many countries had not passed relevant legislation, and many had no policies in place relating to the provision of assistive technology. This suggests that the supply of AT is a relatively low priority for several regions [6]. Lack of Services AT services are frequently in limited supply and located far from where disabled people live. As per the survey [6], many countries say that they had not started programmes to provide AT. Non-governmental organization rarely have the financial or human resources to create country-wide, long-term service delivery systems. Due to their limited geographic reach, their services are generally designed for specific types of AT or impairments. The current system of service delivery is unfair. Inequities have been seen among people with different disabilities, genders, ages, languages and cultures, as well as between people living in different countries or sections of a country or under different economic conditions. When compared to adults, children are less likely to use AT. In addition to a lack of financial resources, it is culturally hard for women in some areas to receive AT because services are operated solely by men. Lack of Products In many nations, assistive products are either not produced at all or are produced on a minimal basis. It is limited not only in terms of quantity but also in terms of product variety, including types, models and sizes. Production of assistive products can be limited by the lack of access to the materials and equipment required. Production might also be limited by market-related reasons. Limited demand results from a lack of awareness of AT or purchasing capability. As a result, there are limited incentives to participate in the production. Local manufacture may not be cost-effective where local markets are tiny and ATrelated duties and import taxes may deter local enterprises from importing materials and equipment.

Assistive Technology

Advances in Data Science-Driven Technologies 201

Although there are many different sorts of ATD accessible around the world, they aren't available everywhere and not every design is appropriate for every situation. As a result, product research and development will continue to be required. There will continue to be a low demand for ATD until the design matches a disabled person's and family's requirements and preferences as well as being appropriate in their physical, social and cultural surroundings. Lack of Inaccessible Environments AT is hampered by behaviors that are either physically or cognitively unavailable. Inaccessible transportation networks or service centers, for example, prohibit disabled people from easily accessing the services and products they require. Physical barriers include passageways and dim lighting, while cognitive barriers include unclear writing and difficult-to-understand symbols. In addition, regardless of the cost or availability of a wheelchair, disabled person will be unable to use it in an inaccessible home and road, and school barriers are frequently worsened after natural disasters and wars. Lack of Human Resources Another obstacle to AT is a shortage of individuals who are appropriately trained in the production, adoption and delivery of products and services. Many countries claim to have insufficient rehabilitation personnel. Lack of Finance Governments must integrate AT and related services into health and community services as well as finance the provision of ATD and services so that they are available at no cost. As household out-of-pocket spending for ATD can be a substantial barrier to access, federal subsidies are required. Existing cadres of health staff can be quickly trained in-service to support the start-up or strengthening of service provision. This should be done in combination with a long-term strategy for developing and training undergraduate curricula. The focus should be on enhancing economies of scale in local manufacturing and assembly as well as reducing or eliminating tariffs on imports, especially where importing countries lack local manufacturing capacity. Sufficient items should be made available and prescribed and fitted appropriately. Disabled people should receive proper training and follow-up, and societal and environmental obstacles should be addressed.

202 Advances in Data Science-Driven Technologies

Goldena and Thangapriya

Assistive Technology’s Challenges Even having a solid basement, technology remains hard due to a variety of challenges [6]. The key obstacles of AT are shown in Fig. (6).

Fig. (6). Challenges in AT

Challenges in Availability Services and products must be made available to the disabled community in sufficient quantities. Low manufacturing and restricted quality, cost restrictions and a lack of government financing, provisions and human resources are all obstacles to enhancing access to AT in low and middle-income nations. Personnel educated to provide these technologies are in short supply, particularly at the provincial and local levels. The expenses of access are exorbitant in many situations where it is conceivable [7]. Challenges in Accessibility People with disabilities need products and services. Access to highly skilled technicians and suppliers is often limited and varies widely between states, counties, and urban and rural areas. Access to AT is also limited by factors such as expectations, legal limits, stereotyping, autonomy, culture, language and dignity. These accessibility challenges also exist in low- and middle-income countries, according to the World Disability Report and are exacerbated by a lack of policies and standards, negative attitudes of society toward people with disabilities, and lack of medical rehabilitation, vocational training and welfare services [7].

Assistive Technology

Advances in Data Science-Driven Technologies 203

Challenges in Affordability Everyone who needs services and products should be able to afford them. Many legal acts suggest that providing AT devices for disabled people in the workplace, residential adjustments, disabled person toilets and so on is mandatory. To eliminate disparities between genders, disability groups, socioeconomic groups and geographic regions, they should be delivered in an equitable manner [7]. Challenges in Adaptability People with disabilities should have their wants and needs satisfied by adapting and modifying services and goods. They must take into account individual variables (such as health, body structure, body function, capacity, gender, age, ethnicity and preference) as well as environmental factors (for example, physical environment, psychosocial environment, climate and culture). Self-awareness and comfort with technology are also important considerations in this decision. It must be simple to learn and use. Individualization must take into account the needs of the person [7]. Challenges in Acceptability To guarantee that technologies and related services are acceptable, factors such as efficiency, reliability, simplicity, safety, comfort and aesthetics should be considered. Customization is critical. Devices for seniors with cognitive impairments were more readily accepted and employed than those for seniors with physical impairments [7]. Challenges in Quality The quality of the services and products must be satisfactory. In terms of strength, durability, capacity, safety and comfort, product quality can be measured using suitable technical standards or guidelines. Although computer and information technologies have the potential to enhance the capabilities of existing AT, product designers have yet to consult with and meet the diverse preferences and demands of persons with disabilities. Screen design, input device design, complex commands and operational procedures are all quality-related issues [7]. Challenges in Research Furthermore, despite the fact that there is a significant and growing need for AT in low and middle-income countries, there is a shortage of research in these areas, preventing the establishment of evidence-based policy and practice. The majority of research evaluating the usefulness of various forms of ATD comes from highincome environments, which is unsurprising. A lack of high-quality, well-

204 Advances in Data Science-Driven Technologies

Goldena and Thangapriya

designed research in this area has been addressed in several reviews of findings to date. The lack of trustworthy information on effectiveness is a critical gap that must be filled immediately. This is important not only for guiding more efficient resource allocation in high-income countries where such technologies are available, but also for allowing evidence-based decisions in low-income countries. The outcome measures for assessing the impact of assistive technologies should be meaningful not only to the target populations, but also and most crucially, to their family and caretakers. Furthermore, there are few systematic cost studies of ATD for dementia patients and their caretakers. Users and caregivers have yet to be consulted during the research, development and design stages in order to produce products that best suit their physical and social contexts and preferences [7]. Challenges in Policy Implementation Irrespective of gender, age or type of impairment, assistive equipment for disabled individuals must be equally available, accessible and inexpensive. Individual country-level policies that direct execution, political will and suitable government structures to support implementation are all necessary for successful international policy implementation. On the other hand, many states do not effectively execute policies and procedures to ensure that assistive equipment is available. The absence of representatives for blind people among governments' policy-making members is partly to blame for the unavailability of information and communication technology resource centers for people with vision impairment. Policy implementation issues are particularly severe in rural areas in many countries for a variety of reasons, including a small number of providers, insufficient infrastructure and a lack of competent workers for AT device training and maintenance [7]. Challenges in Multisectoral Action Multidisciplinary collaboration should effectively contribute to a holistic approach that strengthens the functional skills and autonomy of all potential ATD beneficiaries. This necessitates a multi-agency strategy. Multisectoral efforts across government and commercial sectors are required for universal designs for ATD, buildings, transportation and information and communication technologies. The national AT policy framework can incorporate multi-sectorial participation, particularly from governments, producers, users and consumers. Housing modifications allow people with disabilities to live independently, with community involvement and support from local authorities. Government agencies, businesses and research organizations have demonstrated successful AT innovation through coordinated knowledge transfer, partnerships and targeted

Assistive Technology

Advances in Data Science-Driven Technologies 205

funding that support training, local research and development and the manufacture of high-quality solutions, all with the involvement and active participation of people with disabilities. Effective multisectoral action governance necessitates leadership capacity across sectors and levels of government as well as the nurturing of champions in many sectors who can agree on common goals [4]. FUTURE DIRECTIONS IN ASSISTIVE TECHNOLOGY As technology advances, more barriers for persons with disabilities will be removed. Individuals with disabilities and their loved ones can benefit greatly from AT, and the future promises exciting possibilities for further improvements. The following sections explore future AT directions in various areas: Cognitive Disability Students in special education will benefit from support from healthcare providers, the government and AT corporations in the future as they develop groundbreaking and new technologies [5]. Continuous assistive technology creation, fair cost and gadget accessibility are some of the aspects that require continuous research and advancements as demand and technology used in the twenty-first century grows. Cause-and-effect interactions might be difficult for students with unique needs. Understanding how actions lead to events, such as how performing a math function leads to the right result or predicting what would happen if a button on an AT device, such as a capability switch, is pressed, is an important element of learning. As a result, development is required to establish a cause-an-effect relationship. Motor Disability Future research using activity monitors to follow prosthetic use is suggested. Few studies on psychological aspects of prosthetics, such as prosthesis embodiment, sensory preference and community attitudes about prosthetic use and utilization, have been conducted. Long-term data on community-based activities, particularly in reference to community participation and isolation, which is a prevalent issue among prosthesis users and has been connected to quality-of-life scores, would be beneficial. More research is required for this. Physical activity monitoring in the community may also help researchers better understand the links between physical activity and other parameters, such as prosthetic socket fit for comfort, function and energy savings. Socket fit is critical for successful rehabilitation and restoration of function and mobility, but there are no methods available to objectively assess socket fit. Most researchers did not include information about the weather, the day of the week, the season or whether or not a walking aid was used. When these other factors are taken into account, it is possible to have a

206 Advances in Data Science-Driven Technologies

Goldena and Thangapriya

clearer knowledge of how an individual's activity fluctuates as well as stronger support for clinical ratings and prosthetists' recommendations. HAR (Human Activity Recognition) is a new domain that allows sensors to provide detailed movement analytics, such as gait symmetry, stability for safe ambulation, stride length, compensatory movements and upper-limb movement analytics, which could provide additional information to clinicians as they plan rehabilitation and exercises for prosthesis users to improve prosthesis functionality. The authors urge that when choosing sensors to track physical activity, sensors that give access to raw data be utilized, since this allows for custom data processing and research replication without the limits of certain manufacturers. Visual Disability People with visual problems require more efficient aesthetics for clear vision. Due to the high cost, AT devices should be reasonable. To minimize the perceived stigma, designers should consider aesthetics in addition to functionality, minimizing the likelihood of gadget abandonment. Increased functionality should be the focus of future studies. Auditory Disability The study of how neural signals in a person's brain can be interpreted by a computer to aid communication is called brain-computer interface research. Some academics are researching how a person who is locked in can manage communication software and write out words only by envisioning the movement of his or her hand by implanting electrodes in the brain's motor cortex. Other researchers are working on creating a prosthetic device that can convert a person's ideas into synthesized words and sentences. Another group is working on a wireless gadget that will track brain activity in response to visual stimuli. As a result, researchers can make use of this new technology to create new algorithms for a variety of disabilities, including hearing loss. The Following are Some of the Most Recent AT Research Openings • Future AT practices should concentrate on maximizing the potential of commonplace gadgets as AT for all students, enabling inclusion and lowering stigma [8]. • Additional research should be done on a global scale to establish the state of knowledge and perspectives on future mobility-assistive technology research and development needs and goals [9]. • There is still much to learn and more study is needed, particularly to gain a

Assistive Technology

Advances in Data Science-Driven Technologies 207

better understanding of AT's long-term utility for kids with reading and writing disabilities [10]. • COOK (Cognitive Orthosis for coOKing) looks to have promise in terms of rehabilitating clients with cognitive disabilities, increasing home safety and reducing the need for human supervision. Future research will need to look into how COOK can be adapted to a larger TBI (Traumatic Brain Injury) population, different surroundings and different consumers [11]. • Establish that rules and evidence-based procedures are needed in this sector of study to promote and qualify user involvement. The findings also highlight the importance of conducting research on all aspects that are necessary to provide people with dementia with relevant, effective and long-lasting AT solutions. Future research must go beyond the design and testing stages to provide more knowledge and evidence-based dissemination and adoption techniques. Furthermore, implementing standards for doing and reporting research could be extremely advantageous in terms of improving study design, degree of evidence and effect of research findings. It would also be extremely useful to the research quality [12]. • Parents noticed favourable changes in their children's behaviour, including their children initiating new behaviours for the first time. Parents rapidly learned how to set up and operate the device and were pleased with its present configuration. The comfort of the wearing harness was to be improved in the future [13]. • SmartAbility's plans for the future include the development of a second application that will recommend assistive technologies for the education sector based on the physical and cognitive abilities of users [14]. • Future recommendations are based on the results of two Israeli pilots that tested the platform in a variety of settings and with a variety of stakeholders. These suggestions include ensuring continuity of care and providing a complete user journey, incorporating shared decision-making and self-assessment features, providing data customization and a holistic approach, developing a market network infrastructure and designing the tool as part of a larger service delivery model design [15]. • Essential to apply a conceptual model, standardised tests and collaborate with the players and their trainers. It is hoped that the baseline reported in this study may be useful in future AT or parasports studies [16]. • Making available national data to researchers with fewer conditions associated with its use as well as stakeholders publishing more of their work to build the literature base for AT information in Malawi, which can eventually contribute to evidence-based programming and policies, are areas for improvement [17]. • In India and Nepal, the local government has increased financial resources, yet this is still insufficient. As a result, government budgets for PWD (Person with Disability) related activities and AT services should be greatly increased. In all

208 Advances in Data Science-Driven Technologies

Goldena and Thangapriya

three nations, AT services are still viewed as charities rather than as a fulfilment of PWDs' rights. This needs to change at both the system and practice levels [18]. • Future functionalities should be customizable to meet the health needs of each user and could include smoke detection and reading aloud capabilities. Daily care robots show promise in aged care, particularly in terms of medicine, health and well-being reminders. fundamental value of co-designing and testing robotics in the contexts for which they were created. With a few modifications, widespread adoption of the Bomy robot could be possible in the future [19]. • The future generation of HCAs (Healthcare Cognitive Assistants) will face a major challenge in improving their cognitive abilities. Cognitive assistants' ultimate goal is to imitate human cognition, which is yet poorly understood [20]. • More research is needed to determine the potential benefits of MWC+AOs (Manual Wheel Chair + Add Ons) and the extent to which they can help WMAD (Wheeled Mobility Assistive Devices) users increase their personal autonomy [21]. • The value of mobility devices for activities and participation can help healthcare workers provide mobility devices in situations where environmental aspects from different contexts must be considered. Stakeholders should be aware of the importance of device and service satisfaction [22]. • Future studies should be undertaken with larger samples and for longer periods of time, improving the capabilities of controlled equipment and devices, to see if the benefits persist over time [23]. • Disabled people's activity recognition could be a promising research topic for the elderly. Physiological signals such as ECG (Electrocardiogram), TEB (Thoracic Electrical Bio-Impedance), EDA (Electro Dermal Activity) and smartphone data can be converted to numeric values and analyzed further [24, 25]. CONCLUSION The various disabilities and the problems they face in the absence of assistance are clearly outlined in this paper. If disabled people are provided the suitable AT, they will be able to complete their everyday tasks independently. The number of persons with disabilities is increasing every day, according to statistics. Everyone has the right to equal respect. They cannot be dismissed simply because they have a disability. Now is the moment to put an end to all negative perceptions about disabled persons. CONSENT FOR PUBLICATION Not applicable.

Assistive Technology

Advances in Data Science-Driven Technologies 209

CONFLICT OF INTEREST The author declares no conflict of interest, financial or otherwise. ACKNOWLEDGEMENT We, the authors, would like to express our thanks to the editors for giving us the opportunity to explore this area of our research interests, which helped us in doing a lot of Research throughout which we attempted to reconnoitre many interesting paradigms related to the assistive technology domain. Finally, we thank God Almighty for His abundant grace and mercy, who made this possible. REFERENCES [1]

Reference: Available from..https://www.atia.org/home/at-resources/what-is-at/

[2]

E-book Camyl Gatchalian , Ontario tech university “Assistive technologies in the 21st century” published by “Pressbooks”.https://www.disabilityexpertsfl.com/blog/assistive-devices-for-disabilty-past-present-and-future

[3]

E-book “Understanding the law & assistive technology originally produced by the Family center on Technology and disability” published by “Center on Technology and Disability”.https: //techandcurr2019.pressbooks.com/chapter/21st-century-assistive-tech/

[4]

E-book “Understanding the law & assistive technology originally produced by the Family center on Technology and disability published by Center on Technology and Disability”. http://www.ctd institute.org/sites/default/files/file_attachments/AT-Laws.pdf

[5]

E-book, Special Education Notes” designed and developed by Arpit Pokhriyal..https://www. specialeducationnotes.co.in/paper1Unit1.htm?i=1

[6]

AT possibilities and Limitations” by “National Council for Special Education..https://www.sess.ie/resources/possibilities-and-limitations#:~:text=Assistive%20 technology%20may%20be%20limited,endurance%20abilities%20of%20the%20pupil

[7]

Assistive technology” by “World Health Organization..https://www.who.int/news-room/fac-sheets/detail/assistive-technology

[8]

Aoife McNicholl, Hannah Casey, Deirdre Desmond, and Pamela Gallagher, The impact of AT use for students with impairments in higher education: a systematic review, pp. 362-376, 2019.

[9]

Saleh Alqahtani, James Joseph, and Brad Dicianno, Stakeholder perspectives on research and development priorities for mobility assistive-technology: a literature review, pp. 362-376, . [http://dx.doi.org/10.1080/17483107.2019.1650300]

[10]

Idor Svensson, and Thomas Nordström, "Effects of AT for students with reading and writing impairments", Published Online, pp. 196-208, .

[11]

Stéphanie Pinard, Design and usability evaluation of COOK, an AT for meal preparation for persons with severe TBI., pp. 687-701, .

[12]

Laila Øksnebjerghttps, "Janet Janbek, Bob Woods and Gunhild Waldemar", AT designed to support self-management of people with dementia: user involvement, dissemination and adoption., 2019.orcid.org/0000-0003-2322-8000

[13]

Elena Kokkoni, and James Cole Galloway, User-centredAT assessment of a portable open-area body weight support system for in-home use, pp. 505-512, 2019.

[14]

Paul Whittington, Huseyin Dogan, Keith Phalp, and Nan Jiang, Detecting physical abilities through

210 Advances in Data Science-Driven Technologies

Goldena and Thangapriya

smartphone sensors: an AT application.. [15]

Moran Ran, and David Banes, Basic principles for the development of an AI-based tool for AT decision making., 2020.

[16]

Gabrielle Rodrigues Alves Teixeira, and Ana Cristina de Jesus Alves, Occupational therapy intervention in paralympic sport: a look at low-cost AT for wheelchair rugby, pp. 432-437, .

[17]

I.D. Ebuenyi, J. Kafumba, E.M. Smith, M.Z. Jamali-Phiri, A. Munthali, and M. MacLachlan, "Empirical research and available data on AT for persons with impairments in Malawi", RE:view.

[18]

Jiban Karki, Access to AT for persons with impairments: a critical review from Nepal, India and Bangladesh., 2021.

[19]

Norina Gasteiger, Ho Seok Ahn, and Christine Fok, Older adults’ experiences and perceptions of living with Bomy, an assistive dailycare robot., .

[20]

"Sarah masud preum, Sirajum munir, Meiyi ma, Mohammad samin yasar, David j. Stone, Ronald illiams and homa alemzadeh, john a. Stankovic", A Review of Cognitive Assistants for Healthcare: Trends,Prospects and Future Directions., 2021.

[21]

Mahsa Khalili, Chelsea Jonathan, Nicole Hocking, and Mike Van Der Loos, "Bjorg Gudjonsdottir & Svandis Bjork Gudmundsdottir", Mobility devices for children with physical impairments: use, satisfaction and impact on participation..

[22]

Bjorg Gudjonsdottir, and Svandis Bjork Gudmundsdottir, "Mariana Midori Sime ,Alexandre Luís Cardoso Bissoli , Daniel Lavino-Júnior , Teodiano Freire Bastos-Filho ,Usability", occupational performance and satisfaction evaluation of a smart environment controlled by infrared oculography by people with severe motor impairments., 2021.

[23]

Mariana Midori Sime, Luís Cardoso Bissoli Alexandre, Daniel Lavino-Júnior, and Teodiano Freire Bastos-Filho, Usability, occupational performance and satisfaction evaluation of a smart environment controlled by infrared oculography by people with severe motor impairments, 2021.

[24]

"Thangapriya and Nancy Jasmine Goldena, Comparative analysis of feature selection methods based on activity recognition", published by “Thiruvalluvar university”..

[25]

"Thangapriya and Nancy Jasmine Goldena, Human Activity Recognition using data mining -A survey", published by “A.P.C Mahalaxmi college for women.

Advances in Data Science-Driven Technologies, 2023, 211-217

211

SUBJECT INDEX disease 78 disorders 33, 192

A Ability, cognitive 198, 207, 208 Accessibility 38, 81, 137, 160 cognitive 81, 137, 160 disabilities flipping 38 Age-related macular degeneration 192 Algorithms 57, 58, 59, 60, 61, 63, 122, 125, 126, 130 deep learning 126 deep neural network 63 Artificial neural network (ANN) 56 Assistive devices 2, 6, 13, 23, 85 adoption 23 electrical 2 electronic 85 ontology of 6, 13 Assistive 11, 16, 18, 84 device system 11 listening system 16, 84 robotic systems 18 Assistive technology 6, 7, 12, 73, 74, 81, 86, 186, 187, 196, 201, 203, 204 applications 6 devices (ATD) 6, 7, 12, 73, 74, 81, 86, 186, 187, 196, 201, 203, 204 for cognitive augmentation 6 Audio technology 198 Autism 76, 91, 93, 94 care skills 94 disorders 91, 93, 94 spectrum disorder (ASD) 76 Automated teller machine (ATM) 38, 39 Automatic sign language 102, 117, 118, 120 recognition (ASLR) 117, 118, 120 translation systems 102

B Braille translator software (BTS) 87 Brain 33, 78, 192

C Camera(s) 48, 50, 52, 53, 59, 60, 62, 67, 91, 92, 121, 122, 125 gadget 53 iPhone 91 monochrome 125 Cerebral palsy (CP) 78, 83, 91, 191, 197 Cognitive 79, 85, 186, 190, 191, 194, 205, 207 assistive technology 190 devices 85 disabilities 79, 186, 190, 194, 205, 207 orthosis 207 processes 191 Communication 26, 28, 79, 117, 118, 206 methods 117 obstacles 79 software 206 systems 118 technology 26, 28 Community 22, 188 based methods 22 integration 188 Computer 33, 89, 118 access aids 89 vision techniques 118 Computing technology 33 Convolutional neural network (CNN) 56, 62, 63, 126, 127, 129 Cytomegalovirus 193

D Damage 28, 76, 191, 193, 196 blood vessel 191 neural 196 peripheral nerve 76 Deaf communities 92, 102, 111, 115

Manoj Kumar M V, Immanuel Azaad Moonesar R.D., Ananth Rao, Pradeep N, Annappa, Sandeep Kautish and Vijayakumar Varadarajan (Eds.) All rights reserved-© 2022 Bentham Science Publishers

212 Advances in Data Science-Driven Technologies

Deep 63, 70 convolution system 63 learning-based system 70 DeepNeural networks 126 Degrees of freedom (DoF) 121 Dementia care 19 Design 160,186 web applications 160 wheelchair 186 Devices 2, 30, 31, 53, 59, 60, 65, 81, 84, 124, 194, 195, 206 durable polymer-thick film 124 eye-control 30 haptic response 59 orthotic 2 portable 65 prosthetic 2, 206 speech-producing 194 translating 81 vision-based 53 voice-producing 31 wireless 84 wireless telecommunications 195 Disabilities 93, 188 dysgraphia 93 mental 188 Disease 2, 74, 76, 77 cognitive 76 disabling 77 heart 74 non-communicable 2 sickle cell 77 Disorders 2, 30, 50, 74, 76, 77, 85, 157, 165, 190, 192 autistic spectrum 85 blood 77 cognitive 77 degenerative 192 disabled Major motor 192 hereditary blood 77 mental 74 neurological 76, 77, 157 neuromuscular 30 non-communicable 2 ocular 165

Kumar M V et al.

ophthalmic 165 Dizziness 77, 138, 144 Dwarfism 75

E Echo device 187 Electrocardiogram 208 Electro dermal activity 208 Electromagnetic energy 84 Electronic communication 90, 199, equipment 90 technology 199 Electronic devices 10, 73, 197 Emergence of assistive technology 4 Environment, social 3 Equipment 64, 65, 83, 188 compact camera 64 medical 83 portable camera 64, 65 telecommunications 188 Eye gazing 99 Eyesight, impaired 153 Eye-tracking systems 32

F Federal communications commission (FCC) 188 Flex sensor technology 124 FM 84, 186, 194 amplification technology 186 hearing systems 194 radio waves 84 Foot sensitivity 170, 171 Function 6, 10, 15, 16, 17, 35, 39, 42, 52, 70, 79, 198, 205 action cameras 52 cognitive 17 Fusion deposition manufacturing (FDM) 171

G Gadgets 10, 22, 206

Subject Index

electronic 10 wearable 22 wireless 206 Generative adversarial network (GAN) 117, 127, 128 Global assistive technology community 51 Gloves, wearable sensing 126 Growth disorder 75

H Hamburg notation system 106 Hansen’s disease (HD) 76 Haptic 139, 184 communication system 184 device 139 Hard-soft technologies 54 Hardware 2, 10, 26, 29, 30, 32, 34, 89, 121, 135, 136 devices 10 systems 26 Headaches 76, 100, 144 Healthcare 6, 205 industry 6 providers 205 Health monitoring 19 Hearing 4, 6, 40, 51, 74, 75, 81, 84, 87, 94, 141, 197, 198 aids 4, 6, 40, 51, 74, 75, 81, 84, 94, 197, 198 devices 84, 87, 141 Hidden markov models (HMM) 123, 125, 126 High technology devices 10 Holistic approach 204, 207 Human activity assistive technology (HAAT) 14, 15 Huntington’s disease 76

I Illnesses 21, 29, 74, 76, 83, 85, 157, 190, 192 chronic 29 mental 76, 85 respiratory 74

Advances in Data Science-Driven Technologies 213

Image sensor 168 Impaired vision 35, 36, 48 Impairments, progressive functional 2 Implanting electrodes 206 Individualized education programs (IEP) 5 Information and communication technology (ICTs) 1, 26, 28, 29, 32, 38, 39, 41, 42, 43, 44, 49 Infrared sensors 125 Injury damage 192 Instruments 86, 184 electronic 86 math 184 Intensity, optical devices measure light 124 Internet of things (IoT) 22, 53, 183

K Key-point detection 63 Kinect 120, 130, 168 depth image 168 sensor 120 sign skeleton 130

L Learning 42, 43, 59, 191, 194 activities 42, 43, 194 connections 59 disorders 191 Legislations, telecommunications 188 Linear discriminant analysis (LDA) 125 Local binary patterns (LBP) 122 Locomotor disability 76 Long short-term memory (LSTM) 126 Low tech assistive technology 9

M Machine learning 117, 125,130, 119 algorithms 125, 130 Machine translation system 115 Matching 14, 15, 130 person and technology (MPT) 14, 15

214 Advances in Data Science-Driven Technologies

techniques 130 Mathematical assistive technology 197 Mechanism 6, 34, 38, 67, 129, 130, 142, 196 gesture-based learning 129 Medical emergency response system 86 Memory 16, 137, 190, 194 issues 137 loss 16 problems 190, 194 Mental 2, 16, 75, 76, 189 dysfunctions 189 health disorder 2, 76 process 16 retardation 75 Methods 53, 56, 122, 135 braille display 135 deep learning-based 56, 122 sensor 53 Microphone 4, 20, 84, 89, 125, 139, 194, 198 headsets 20 multi-array 125 Microsoft active accessibility (MSAA) 135, 159 Mid-technology devices 8 Mobile 19, 49, 53, 167, 186 applications 186 devices 19, 49, 53, 167 Mobility 16, 18, 27, 73, 79, 80, 81, 82, 83, 87, 84, 89, 90, 92, 158, 165, 166, 172, 174, 191, 194, 197, 208 devices 82, 83, 87, 197, 208 disabilities 79, 80, 83, 158 obstruct 79 restricted 27 Motion 52, 121, 127, 129, 148, 191 sensors 52 Motivation of sign language recognition 123 MT system 113, 114 prototype 113 Multiple sclerosis (MS) 77 Muscular dystrophy 76, 83

Kumar M V et al.

N Neural 108, 123, 125, 126, 127, 129 machine translation (NMT) 127 networks (NN) 108, 123, 125, 126, 127, 129 Neurological diseases 76, 85 Non-assistive computer systems 35 Numerous electronic assistance systems 53

O Object detection system 50 OCR 58, 60, 65, 67 detection 65 software 67 technique 58, 60 Ontological matching initiatives 14 Ontology 1, 14, 15 based assistive devices 1 of assistive technology 14, 15 Optical 19, 35, 49, 58, 60, 61, 65, 67, 68, 69, 86, 90 character recognition (OCR) 19, 35, 49, 58, 60, 61, 65, 67, 68, 69, 90 magnifiers 86 Optics technology 124

P Parkinson’s disease (PD) 76, 78, 192 movement 192 Personal 19, 55, 86, 186 Assistance services 55 emergency response systems (PERS) 19, 86, 186 Photosensitive epilepsy 144 Physical therapist 74 Placeholder textual content 146 Planet health organization (PHO) 32 Plan rehabilitation 206 Platform, railway 165 Psychological distress 3 Puff technology 185

Subject Index

Pulse-coupled neural networks (PCNN) 123 PWD’s learning activities 44

Q QR codes 53, 167

R Recognition accuracy, real-time 64 Recursive neural network (RNN) 56, 117, 127, 129 Rehabilitation act 4, 188 Relational autonomy 19 Repetitive stress injury (RSI) 147 Research on sign language recognition 117 Resources 5, 29, 54, 57, 58, 160, 161, 171 educational 5 lower computational 57 Respiratory issues 54 Robotic systems 58

S Schizophrenia 136, 157 Self-organizing feature maps (SOFM) 125 Sensors 19, 48, 50, 121, 122, 124, 126, 130, 167, 168, 186, 206 abduction 121 electronic 186 flex 124 sensitive 124 ultrasonic 167 water 167 Sensory 26, 153, 167 impairment 26 substitution system 167 Serotek system access 153 Service(s) 2, 41, 188 delivery systems 2 telecommunication 188 web 41 SIFT technique 57 Signals 39, 84, 179, 206

Advances in Data Science-Driven Technologies 215

electrical 84 haptic 179 neural 206 visual 39 Sign 120, 129 generation process 120 recognition accuracy 129 Sign language(s) 40, 75, 99, 101, 102, 103, 104, 117, 118, 119, 120, 127, 128, 129, 130 generation 117, 118, 120, 127 notation system 120 production (SLP) 117, 119, 127, 129, 130 transcription 120 translation systems 102, 103 Sign language recognition (SLR) 117, 118, 119, 120, 123, 126, 127, 129, 130 technique 130 SLR glove systems 121 Snellen test 169 Social management system 19 Software 10, 26, 29, 31, 32, 33, 34, 35, 37, 53, 54, 135, 136, 137, 138, 139, 156, 161, 196 academic tutorial 34 commercial 161 packages 35 platforms 26 program 139, 156 screen magnification 32, 33 Sounds 3, 5, 6, 28, 75, 84, 103, 165, 185, 192, 197 visualization 5 Space technology 197 Spatial-aware mobility 16 Speech 10, 18, 86, 118, 195 and voice recognition devices 86 generating devices (SGDs) 10, 18, 118 recognition system 195 Squeeze machine 85 Support vector machines (SVM) 56, 123, 126 Syndrome, carpal tunnel 3, 34

216 Advances in Data Science-Driven Technologies

T

Kumar M V et al.

U Ultrasonic distance measurements 167

Tactile 164, 165, 169, 171, 197 surface indicators (TSIs) 164, 165, 169 techniques 197 test 171 Techniques 51, 52, 55, 56, 57, 58, 59, 60, 117, 118, 119, 120, 124, 125, 126, 127, 129, 168 central scene object detection 168 clustering 125 machine learning 117, 119, 120, 129 Technologies 22, 50, 59, 167, 168, electronic 22 smartphone 22 ultrasonic 167, 168 vision-based 59 wireless 50 TGSI installations 169 Thalassemia 77 Thermoplastic polyurethane (TPU) 164, 171 Thoracic electrical bio-impedance 208 Tiles 164, 165, 171, 172, 173, 175, 176, 177 adhesive 165 slippery 177 Tissue integrity 16 Toggle keys 32 Toileting equipments 89 Tools 9, 137, 196 digital 9 testing 137 visual tracking 196 Tractability 137, 159 issues 137 ratings 159 Traditional technology 3 Translation 86, 98, 103, 105, 114, 115 automatic 115 system 86, 98, 103, 105, 114 Traumatic brain injury 207 Tritanopia disorder 145 Typography 137, 138, 140, 158

V Vibration motors 167 Video(s) 3, 128 generating 128 network 128 on-screen 3 Viral infectious hearing loss 193 Vision 28, 32, 49, 67, 75, 77, 79, 87, 86, 89, 90, 117, 120, 121, 122, 136, 140, 164, 165, 167, 169, 170, 183, 192, 204 based system 122 blurred 77 disabilities 79, 192 impairment 28, 183, 204 improvement 167 loss 32, 49, 75, 87, 169, 192 reduced 67 Visual 75, 87, 192, 196, 197, 206 disability 87, 192, 196, 197, 206 gesture language 75 Voice 20, 30, 35, 40, 90, 120, 123, 135, 140, 157, 196 computer-generated 196 computerized 20 Voltage 124, 125, 126, 172 fluctuation 124 signal conditions 125 Volunteers 126, 172

W WCAG compliance 136 Wearable technologies 22 Web applications 154, 161 Wheelchairs 4, 18, 19, 74, 81, 82, 83, 89, 94, 197, 199, 201 powered 82, 83 sports 83 target 83

Subject Index

Wireless communication 84 Workplace assistive technology 20

Advances in Data Science-Driven Technologies 217