EUVIMEDEUVIMED entdecken
Crossref

Advancing Bangla Sign Language Detection through Dataset Creation, Model Comparison, and Deployment

Bristy Chakraborty, Md. Masudur Rahman, Apurba Adhikary, Minoru W. Yoshida

International Journal of Engineering and Manufacturing · 08.06.2026 · 16 (3) · 11-29

DOI: 10.5815/ijem.2026.03.02 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Engineering and Manufacturing, erschienen am 08.06.2026. Verfasst von Bristy Chakraborty und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Bangla Sign Language is a unique sign language. Due to a lack of interpreters, the hearing- andspeechimpaired community face challenges while communicating with the broader community. Recent studies have beencon-ducted to reduce the gap between these two communities. But most of the researchers used a dataset with a controlledenvironment. We know the performance of a system highly depends on dataset quality. In this paper, we have createdanew dataset, “BanglaSignSet” including 46 unique signs with over 10k images. We have carefully annotated andlabeledthe images using Roboflow. Our proposed dataset, “BanglaSignSet” consists of images with high resolution, goodquality, and adequate variation in environment and person. The constructed dataset has been trained using the most recent deep learning model, such as YOLOv8. We have also implemented different versions of the YOLOv8 model, suchasYOLOv8n, YOLOv8s, and YOLOv8m. Additionally, we evaluated EfficientNet-B0 as a classification-based baselinetobroaden the experimental comparison. The performance of models has been measured using different evaluationmetricssuch as mAP, precision, recall, and f1 score. A comparative analysis has been conducted based on the performanceofthe model. By comparative analysis we found a well-suited model, YOLOv8n, to deploy into a web-based application. To find the suitable model to deploy, we have considered factors such as memory requirement and inference speed. Wehave integrated the YOLOv8n model into a web application using the Python language. We have also testedthewebapplication on Android devices and laptops. The web application detects signs from image input successfully

Zitieren

APA

Chakraborty, B., Rahman, M. M., Adhikary, A., & Yoshida, M. W. (2026). Advancing Bangla Sign Language Detection through Dataset Creation, Model Comparison, and Deployment. International Journal of Engineering and Manufacturing, 16(3), 11-29. https://doi.org/10.5815/ijem.2026.03.02

Vancouver

Chakraborty B, Rahman MM, Adhikary A, Yoshida MW. Advancing Bangla Sign Language Detection through Dataset Creation, Model Comparison, and Deployment. International Journal of Engineering and Manufacturing. 2026;16(3):11-29. doi:10.5815/ijem.2026.03.02

BibTeX
@article{Chakraborty2026,
  author = {Chakraborty, Bristy and Rahman, Md. Masudur and Adhikary, Apurba and Yoshida, Minoru W.},
  title = {Advancing Bangla Sign Language Detection through Dataset Creation, Model Comparison, and Deployment},
  journal = {International Journal of Engineering and Manufacturing},
  year = {2026},
  volume = {16},
  number = {3},
  pages = {11--29},
  doi = {10.5815/ijem.2026.03.02}
}
Kostenlos starten →