Development of Malaysian sign language Recognition system based on deep learning approach

Bee, Wei Hou (2025) Development of Malaysian sign language Recognition system based on deep learning approach. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

Text (Full Text) Development of Malaysian sign language Recognition system based on deep learning approach.pdf - Submitted Version Download (8MB)

Abstract

In Malaysia, Malaysian Sign Language (MSL) serves as a vital means of communication for the deaf community, yet its usage in daily interactions remains limited, particularly among hearing individuals. The absence of understanding MSL poses significant inconvenience and barriers for effective communication between deaf and hearing individuals. Previous research predominantly focused on image classification, constraining the potential of MSL translation systems. This project aims to develop an advanced MSL translator capable of detecting various sign gestures from both images and videos and translating them into coherent sentences and speech. Leveraging the power of OpenCV, TensorFlow, and Mediapipe software, the proposed model offer real-time translation capabilities, facilitating seamless communication between the deaf and hearing communities. By enabling the deaf to communicate effectively with hearing individuals, whether in face-to-face interactions or online meetings, the developed MSL translator endeavors to bridge the longstanding communication gap. Its successful implementation holds the promise of fostering inclusivity, understanding, and collaboration between these two communities, there by enhancing social integration and accessibility for the deaf population in Malaysia. In conclusion, the realization of this translator signifies a crucial step towards breaking down communication barriers, fostering empathy, and promoting inclusivity in Malaysian society.

Actions (login required)

View Item