Abstract
Sign language is used to communicate using hand movements rather than words or written language. Typically, this approach is useful for the deaf or mute people since they cannot utilize other forms of communication due to their speech and hearing impairments. The majority of Bangladeshi people are unfamiliar with sign language. As a consequence, deaf or mute people are unable to communicate with general individuals. Therefore, to address this issue, computer vision and supervised learning techniques are utilized to recognize images of the Bengali Sign generated with both hands. Thus, in this research, we proposed a method to evaluate the performance of our own dataset named BdSL 49 which contains 49 classes and 29,428 images. We use our dataset to train the latest transfer learning benchmark models such as Xception, InceptionV3, InceptionResNetV2, MobileNet, MobileNetV2, ResNet50V2, and ResNet101V2 which are used for sign image recognition. The experimental result analysis of the benchmark models has been done where the performance of the models is satisfactory. Among the seven models, the Xception, InceptionResnetV2, and MobileNet achieved the highest F1-score of 93, 91, and 92%, respectively. Additionally, we compared our dataset with the state-of-the-art dataset. On the basis of the model’s performance, we can conclude that our dataset is pretty standard. Finally, the paper is concluded with some future directions.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Disability in Bangladesh. https://en.wikipedia.org/wiki/Disability_in_Bangladesh. Accessed on September 1, 2022
Hasib A, Khan SS, Eva JF, Khatun M, Haque A, Shahrin N, Rahman R, Murad H, Islam M, Hussein MR et al (2022) Bdsl 49: a comprehensive dataset of Bangla sign language. arXiv preprint arXiv:2208.06827
Islam MS, Mousumi SSS, Jessan NA, Rabby ASA, Hossain SA (2018) Ishara-lipi: the first complete multipurpose open access dataset of isolated characters for Bangla sign language. In: 2018 international conference on Bangla speech and language processing (ICBSLP). IEEE, pp 1–4
Islalm MS, Rahman MM, Rahman MH, Arifuzzaman M, Sassi R, Aktaruzzaman M (2019) Recognition Bangla sign language using convolutional neural network. In: 2019 international conference on innovation and intelligence for informatics, computing, and technologies (3ICT). IEEE, pp 1–6
Podder KK, Chowdhury ME, Tahir AM, Mahbub ZB, Khandakar A, Hossain MS, Kadir MA (2022) Bangla sign language (bdsl) alphabets and numerals classification using a deep learning model. Sensors 22(2):574
Jim AAJ, Mendeley (2021) KU-BdSL: Khulna University Bengali sign language dataset. https://doi.org/10.17632/SCPVM2NBKM.1
Rafi AM, Nawal N, Bayev NSN, Nima L, Shahnaz C, Fattah SA (2019) Image-based Bengali sign language alphabet recognition for deaf and dumb community. In: 2019 IEEE global humanitarian technology conference (GHTC). IEEE, pp 1–7
Angona TM, Shaon AS, Niloy KTR, Karim T, Tasnim Z, Reza SS, Mahbub TN (2020) Automated Bangla sign language translation system for alphabets by means of mobilenet. TELKOMNIKA (Telecommun Comput Electron Control) 18(3):1292–1301
Khatun A, Shahriar MS, Hasan MH, Das K, Ahmed S, Islam MS (2021) A systematic review on the chronological development of Bangla sign language recognition systems. In: 2021 Joint 10th international conference on informatics, electronics & vision (ICIEV) and 2021 5th international conference on imaging, vision & pattern recognition (icIVPR). IEEE, pp 1–9
Youme SK, Chowdhury TA, Ahamed H, Abid MS, Chowdhury L, Mohammed N (2021) Generalization of Bangla sign language recognition using angular loss functions. IEEE Access 9:165351–165365
Acknowledgements
This work is supported by the Institute of Energy, Environment, Research, and Development (IEERD), University of Asia Pacific (UAP), Bangladesh.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Khan, S.S., Haque, A., Khatun, N., Begum, N., Jahan, N., Helaly, T. (2023). An Evaluation of BdSL 49 Dataset Using Transfer Learning Techniques: A Review. In: Kaiser, M.S., Waheed, S., Bandyopadhyay, A., Mahmud, M., Ray, K. (eds) Proceedings of the Fourth International Conference on Trends in Computational and Cognitive Engineering. Lecture Notes in Networks and Systems, vol 618. Springer, Singapore. https://doi.org/10.1007/978-981-19-9483-8_37
Download citation
DOI: https://doi.org/10.1007/978-981-19-9483-8_37
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-9482-1
Online ISBN: 978-981-19-9483-8
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)