Attention-Based Approach for Arabic Sign Language Recognition, Supporting Differently Abled Persons

Sign language serves as a critical communication channel for non-speaking individuals. In this paper, we use a vision transformer (ViT) to classify static images representing the alphabets of Arabic Sign Language. We employ two Kaggle Arabic alphabet datasets, one with over 15,000 static images, und...

Full description

Saved in:
Bibliographic Details
Published inJournal of Disability Research Vol. 4; no. 4
Main Authors Almufareh, Maram Fahaad, Tehsin, Samabia, Humayun, Mamoona, Kausar, Sumaira, Farooq, Asad
Format Journal Article
LanguageEnglish
Published 04.08.2025
Online AccessGet full text

Cover

Abstract Sign language serves as a critical communication channel for non-speaking individuals. In this paper, we use a vision transformer (ViT) to classify static images representing the alphabets of Arabic Sign Language. We employ two Kaggle Arabic alphabet datasets, one with over 15,000 static images, under varying background conditions, consisting of 31 classes. Our approach emphasizes the inclusivity aspect by aiming to minimize background-related biases and enhance accessibility for persons with a communication disability. Experiments reveal that ViT outperforms traditional convolutional neural networks by achieving a peak accuracy. Cross-data validation results show the reliability, generalizability, and robustness of findings by testing results across multiple, independent datasets.
AbstractList Sign language serves as a critical communication channel for non-speaking individuals. In this paper, we use a vision transformer (ViT) to classify static images representing the alphabets of Arabic Sign Language. We employ two Kaggle Arabic alphabet datasets, one with over 15,000 static images, under varying background conditions, consisting of 31 classes. Our approach emphasizes the inclusivity aspect by aiming to minimize background-related biases and enhance accessibility for persons with a communication disability. Experiments reveal that ViT outperforms traditional convolutional neural networks by achieving a peak accuracy. Cross-data validation results show the reliability, generalizability, and robustness of findings by testing results across multiple, independent datasets.
Author Almufareh, Maram Fahaad
Kausar, Sumaira
Farooq, Asad
Tehsin, Samabia
Humayun, Mamoona
Author_xml – sequence: 1
  givenname: Maram Fahaad
  orcidid: 0000-0002-6613-0831
  surname: Almufareh
  fullname: Almufareh, Maram Fahaad
– sequence: 2
  givenname: Samabia
  orcidid: 0000-0003-0761-0752
  surname: Tehsin
  fullname: Tehsin, Samabia
– sequence: 3
  givenname: Mamoona
  orcidid: 0000-0001-6339-2257
  surname: Humayun
  fullname: Humayun, Mamoona
– sequence: 4
  givenname: Sumaira
  orcidid: 0000-0003-4229-7802
  surname: Kausar
  fullname: Kausar, Sumaira
– sequence: 5
  givenname: Asad
  orcidid: 0009-0005-5090-159X
  surname: Farooq
  fullname: Farooq, Asad
BookMark eNotkMlOwzAARC1UJErplbM_ABcv9ZJjaFlVCdTCObIdOxgVO7LTQ_-eFDjNZeZJ8y7BJKboALgmeMElqeTty3qLKKYcYa7EGZhSIQWigrEJmBLBFaoqQi_AvJQvjDFjZIklnwJbD4OLQ0gR3eniWlj3fU7afkKfMqyzNsHCXegi3OjYHXTn4NbZ1MVw2tzA3aHvUx5C7OA6eO_yCNsfYW32I-vN5ZJiuQLnXu-Lm__nDHw83L-vntDm9fF5VW-QJdVSIKeWxitriMNGtkaLlijWMu-M1qql2jKhCR6vVEoTxiUltJWKKM2rseYcm4HFH9fmVEp2vulz-Nb52BDc_FpqRkvNyVJzssR-AFDeXR8
Cites_doi 10.1016/j.array.2022.100141
10.1155/2022/4567989
10.1109/SCEECS.2018.8546967
10.1109/ICAEEE.2018.8642983
10.14569/IJACSA.2018.090327
10.1109/ICME.2015.7177428
10.48550/arXiv.2010.11929
10.5120/14106-2145
10.5815/ijigsp.2018.08.03
10.1016/j.dib.2021.107021
10.3390/s24113683
10.1109/MVT.2021.3140047
10.1109/ICCVW.2011.6130290
10.19026/rjaset.15.5292
10.1109/INISTA.2011.5946072
10.11591/ijeecs.v22.i2.pp1096-1107
10.1145/3373625.3417023
10.1007/s12652-020-01790-w
10.4236/jilsa.2012.41004
10.48550/arXiv.2008.09918
10.1016/j.neucom.2020.12.006
10.48550/arXiv.2305.14527
10.1109/ICICCS56967.2023.10142596
ContentType Journal Article
DBID AAYXX
CITATION
DOI 10.57197/JDR-2025-0586
DatabaseName CrossRef
DatabaseTitle CrossRef
DatabaseTitleList CrossRef
DeliveryMethod fulltext_linktorsrc
EISSN 2676-2633
ExternalDocumentID 10_57197_JDR_2025_0586
GroupedDBID AAYXX
ABDBF
ALMA_UNASSIGNED_HOLDINGS
CITATION
ESX
GROUPED_DOAJ
ID FETCH-LOGICAL-c1946-e84bf8cb1e0b7dba6d183d3febaa8d2ac36a1065898a1357212d7818a593d3ee3
ISSN 1658-9912
IngestDate Thu Aug 14 00:01:00 EDT 2025
IsDoiOpenAccess false
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 4
Language English
License https://creativecommons.org/licenses/by/4.0
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c1946-e84bf8cb1e0b7dba6d183d3febaa8d2ac36a1065898a1357212d7818a593d3ee3
ORCID 0000-0003-4229-7802
0000-0002-6613-0831
0000-0003-0761-0752
0000-0001-6339-2257
0009-0005-5090-159X
OpenAccessLink https://www.scienceopen.com/hosted-document?doi=10.57197/JDR-2025-0586
ParticipantIDs crossref_primary_10_57197_JDR_2025_0586
PublicationCentury 2000
PublicationDate 2025-08-04
PublicationDateYYYYMMDD 2025-08-04
PublicationDate_xml – month: 08
  year: 2025
  text: 2025-08-04
  day: 04
PublicationDecade 2020
PublicationTitle Journal of Disability Research
PublicationYear 2025
References Wikipedia Contributors (r27) 2025
NC Camgoz (r5) 2020
R Alzohairi (r4) 2018; 9
World Health Organization (r28) 2021
A Alani (r2) 2021; 22
S Katoch (r16) 2022; 14
B Surya (r10) 2023
F Işikdoğan (r14) 2011
M Maraqa (r19) 2012; 4
S Nagarajan (r21) 2013; 82
M Albarham (r3) 2023
A Dosovitskiy (r6) 2021
R El Kharoua (r7) 2023
A Imran (r13) 2021; 36
RD Raj (r24) 2018
Z Hu (r11) 2022; 17
M Zakariah (r29) 2022; 2022
J Huang (r12) 2015
S Shivashankara (r25) 2018; 10
C Gleason (r9) 2020
M Mustafa (r20) 2021; 12
A Kapitanov (r15) 2023
Abdel-Samie (r1) 2018; 15
O Koller (r17) 2020
TH Noor (r22) 2024; 24
L Gao (r8) 2021; 434
N Pugeault (r23) 2011
I Mahmud (r18) 2018
References_xml – volume: 14
  year: 2022
  ident: r16
  article-title: Indian Sign Language recognition system using SURF with SVM and CNN
  publication-title: Array
  doi: 10.1016/j.array.2022.100141
– volume: 2022
  year: 2022
  ident: r29
  article-title: Sign language recognition for Arabic alphabets using transfer learning technique
  publication-title: Comput Intell Neurosci
  doi: 10.1155/2022/4567989
– start-page: 1
  year: 2018
  ident: r24
  article-title: British Sign Language recognition using HOG
  doi: 10.1109/SCEECS.2018.8546967
– start-page: 1
  year: 2018
  ident: r18
  article-title: Efficient noise reduction and HOG feature extraction for sign language recognition
  doi: 10.1109/ICAEEE.2018.8642983
– volume: 9
  start-page: 1
  issue: 3
  year: 2018
  ident: r4
  article-title: Image based Arabic Sign Language recognition system
  publication-title: Int J Adv Comput Sci Appl
  doi: 10.14569/IJACSA.2018.090327
– start-page: 1
  year: 2015
  ident: r12
  article-title: Sign language recognition using 3D convolutional neural networks
  doi: 10.1109/ICME.2015.7177428
– year: 2021
  ident: r6
  article-title: An image is worth 16×16 words: Transformers for image recognition at scale
  doi: 10.48550/arXiv.2010.11929
– volume: 82
  start-page: 28
  year: 2013
  ident: r21
  article-title: Static hand gesture recognition for sign language alphabets using edge oriented histogram and multi class SVM
  publication-title: Int J Comput Appl
  doi: 10.5120/14106-2145
– volume: 10
  start-page: 18
  issue: 8
  year: 2018
  ident: r25
  article-title: American Sign Language recognition system: An optimal approach
  publication-title: Int J Image Graph Signal Process
  doi: 10.5815/ijigsp.2018.08.03
– volume-title: SignWriting
  year: 2025
  ident: r27
– volume-title: Deafness and hearing loss
  year: 2021
  ident: r28
– volume: 36
  year: 2021
  ident: r13
  article-title: Dataset of Pakistan Sign Language and automatic recognition of hand configuration of Urdu alphabet through machine learning
  publication-title: Data Brief
  doi: 10.1016/j.dib.2021.107021
– volume: 24
  issue: 11
  year: 2024
  ident: r22
  article-title: Real-time Arabic Sign Language recognition using a hybrid deep learning model
  publication-title: Sensors
  doi: 10.3390/s24113683
– volume: 17
  start-page: 57
  issue: 4
  year: 2022
  ident: r11
  article-title: Toward human-centered automated driving: A novel spatiotemporal vision transformer-enabled head tracker
  publication-title: IEEE Veh Technol Mag
  doi: 10.1109/MVT.2021.3140047
– start-page: 1114
  year: 2011
  ident: r23
  article-title: Spelling it out: Real-time ASL fingerspelling recognition
  doi: 10.1109/ICCVW.2011.6130290
– volume-title: RGB Arabic Alphabets Sign Language Dataset
  year: 2023
  ident: r3
– volume: 15
  start-page: 57
  year: 2018
  ident: r1
  article-title: Arabic Sign Language recognition using Kinect sensor
  publication-title: Res J Appl Sci Eng Technol
  doi: 10.19026/rjaset.15.5292
– start-page: 10020
  year: 2020
  ident: r5
  article-title: Sign language transformers: Joint end-to-end sign language recognition and translation
– start-page: 264
  year: 2011
  ident: r14
  article-title: Automatic recognition of Turkish
  doi: 10.1109/INISTA.2011.5946072
– volume: 22
  start-page: 1096
  year: 2021
  ident: r2
  article-title: ArSL-CNN: A convolutional neural network for Arabic Sign Language gesture recognition
  publication-title: Indones J Electr Eng Comput Sci
  doi: 10.11591/ijeecs.v22.i2.pp1096-1107
– start-page: 1
  year: 2020
  ident: r9
  article-title: Disability and the COVID-19 pandemic: using twitter to understand accessibility during rapid societal transition
  doi: 10.1145/3373625.3417023
– volume: 12
  start-page: 410
  year: 2021
  ident: r20
  article-title: RETRACTED ARTICLE: A study on Arabic Sign Language recognition for differently abled using advanced machine learning classifiers
  publication-title: J Ambient Intell Human Comput
  doi: 10.1007/s12652-020-01790-w
– volume: 4
  start-page: 41
  year: 2012
  ident: r19
  article-title: Recognition of Arabic Sign Language (ArSL) using recurrent neural networks
  publication-title: J Intell Learn Syst Appl
  doi: 10.4236/jilsa.2012.41004
– year: 2020
  ident: r17
  article-title: Quantitative survey of the state of the art in sign language recognition
  doi: 10.48550/arXiv.2008.09918
– volume: 434
  start-page: 45
  year: 2021
  ident: r8
  article-title: RNN-transducer based Chinese Sign Language recognition
  publication-title: Neurocomputing
  doi: 10.1016/j.neucom.2020.12.006
– volume-title: No background RGB Arabic Alphabets Sign Language
  year: 2023
  ident: r7
– year: 2023
  ident: r15
  article-title: Slovo: Russian Sign Language dataset
  doi: 10.48550/arXiv.2305.14527
– volume-title: An efficient real-time Indian sign language (ISL) detection using deep learning
  year: 2023
  ident: r10
  doi: 10.1109/ICICCS56967.2023.10142596
SSID ssj0003314075
Score 2.2992408
Snippet Sign language serves as a critical communication channel for non-speaking individuals. In this paper, we use a vision transformer (ViT) to classify static...
SourceID crossref
SourceType Index Database
Title Attention-Based Approach for Arabic Sign Language Recognition, Supporting Differently Abled Persons
Volume 4
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1Lb9QwELaWcuGCQIB4FfmAxGExrPNwkuPCUlXlIcS2Um_R-KWttA-UTQ7tL-BnM7azqakWqXCJIseeWJlP42-cmTEhr4VB2mkrw3SaTVhmqpRVqsyZLSqpdcKtTVyC89dv4vgsOznPz0ejX1HUUtfKd-pqb17J_2gV21CvLkv2HzQ7CMUGvEf94hU1jNdb6XjatiFakX3AxUg7ShkSpFzs4LQBeaHGc3fw5pd-V9KxxBAvFH62uzM9N42PfJ71J6W0y8vxVC5R2ndPxrd_oa-zvjivJ_HRlpgDz3LVWWjMok8GgtX4CBYAetgmMIttqF0whxVKgQhccNmF8GJYbTbr4cln6LYhFnyOfS4aiLcrktwHy2WRhUXKw5CUBhNsfFsiCsESESpi7MxyFqEv22fs84L7KsInsx8svCnfV1X7xmo3xCCi9-Ml1Di-duNrN_4OuZugw5FEzrlb09MUHVFftXmYf6gA6kW8_2MKEcOJqMrpA3K_VxKdBsA8JCOzfkTUDbDQHVgogoUGsFAHFroDC43A8pZeQ4VGUKEeKrSHymNydvTp9OMx60_YYIpXmWCmzKQtleRmIgstQWi08Dq1RgKUOgGVCuCOpFYl8DQvkOfoAike5BV2MyZ9Qg7Wm7V5SijXkwlYZQ0u_kixDQhZKEg5txoUur3PyJvdV6l_hkIq9X4VPL91zxfk3jXEXpKDtunMIXLEVr7y6vsNcnxpGg
linkProvider Directory of Open Access Journals
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Attention-Based+Approach+for+Arabic+Sign+Language+Recognition%2C+Supporting+Differently+Abled+Persons&rft.jtitle=Journal+of+Disability+Research&rft.au=Almufareh%2C+Maram+Fahaad&rft.au=Tehsin%2C+Samabia&rft.au=Humayun%2C+Mamoona&rft.au=Kausar%2C+Sumaira&rft.date=2025-08-04&rft.issn=1658-9912&rft.eissn=2676-2633&rft.volume=4&rft.issue=4&rft_id=info:doi/10.57197%2FJDR-2025-0586&rft.externalDBID=n%2Fa&rft.externalDocID=10_57197_JDR_2025_0586
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1658-9912&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1658-9912&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1658-9912&client=summon