Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Jun 18:49:109329.
doi: 10.1016/j.dib.2023.109329. eCollection 2023 Aug.

BDSL 49: A comprehensive dataset of Bangla sign language

Ayman Hasib  1 , Jannatul Ferdous Eva  1 , Saqib Sizan Khan  1 , Mst Nipa Khatun  1 , Ashraful Haque  1 , Nishat Shahrin  1 , Rashik Rahman  1 , Hasan Murad  2 , Md Rajibul Islam  3 , Molla Rashied Hussein  1
Affiliations

BDSL 49: A comprehensive dataset of Bangla sign language

Ayman Hasib et al. Data Brief. .

Abstract

Language is a method by which individuals express their thoughts. Each language has its own alphabet and numbers. Oral and written communication are both effective means of human interaction. However, each language has a sign language equivalent. Hearing-impaired and/or nonverbal individuals communicate through sign language. BDSL is the abbreviation for the Bangla sign language. The dataset contains images of hand signs in Bangla. The collection comprises 49 individual sign language images of the Bengali alphabet. BDSL49 is a set of 29,490 images with 49 labels. During data collection, images of fourteen distinct adults, each with a unique appearance and context, were captured. During data preparation, numerous strategies have been utilized to reduce noise. This dataset is available for free to researchers. Using techniques such as machine learning, computer vision, and deep learning, they are able to develop automated systems. Moreover, two models were applied to this dataset. The first is for detection, and the second is for identification.

Keywords: Bangla language; Computer vision; Deep learning; Hand configuration; Natural language processing; Sign language; YOLOv4 mode.

PubMed Disclaimer

Conflict of interest statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Figures

Fig 1:
Fig. 1
The dataset creation process.
Fig 2:
Fig. 2
Dataset folders format in data repository.
Fig 3:
Fig. 3
Annotation process.
Fig 4:
Fig. 4
Accuracy comparison of various pre-trained models on the recognition dataset.
Fig 5:
Fig. 5
Mean-average precision chart of the YOLOv4 model on the detection dataset
See this image and copyright information in PMC

References

    1. Alauddin M, Joarder AH. In: Hearing Impairment: An Invisible Disability How You Can Live With a Hearing Impairment. Suzuki JI, Kobayashi T, Koga K, editors. Springer; Japan: 2004. Deafness in Bangladesh; pp. 64–69. - DOI
    1. M Sanzidul Islam, S Sultana Sharmin Mousumi, Jessan NA, AKM Shahariar Azad Rabby, Ishara-Lipi Akhter Hossain S. 2018 International Conference on Bangla Speech and Language Processing (ICBSLP) 2018. The first complete multipurposeopen access dataset of isolated characters for Bangla sign language; pp. 1–4. - DOI
    1. Rafi AM, Nawal N, Bayev NSN, Nima L, Shahnaz C, Fattah SA. 2019 IEEE Global Humanitarian Technology Conference (GHTC) 2019. Image-based Bengali sign language alphabet recognition for deaf and dumb community; pp. 1–7. - DOI
    1. Al Jaid Jim A. Khulna University Bengali Sign Language Dataset; 2021. KU-BdSL. Published online.
    1. Hoque OB, Jubair MI, Akash AF, Islam MS. In: Computer Vision – ACCV 2020 Workshops. Sato I, Han B, editors. Springer International Publishing; 2021. BdSL36: a dataset for Bangladeshi sign letters recognition; pp. 71–86.

LinkOut - more resources