Computer Vision-Based Bangla Numerical Sign Language Recognition

Abstract

Sign Language is the mode of communication among the deaf and dumb. However, integrating them into the main stream is very difficult as the majority of the society is unaware of their language. So, to bridge the communication gap between the hearing and speech impaired and the rest in Bangladesh, I conducted a research to recognize Bangla sign language using a computer-vision based approach. Sign language not only help for the people who can't speak or hear, it's also help for human computer interaction system or robotics system. To achieve my goals, I used Convolutional Neural Networks (CNNs) to train individual signs. In the future, this research, besides helping as an interpreter, can also open doors to numerous other applications like sign language tutorials or dictionaries and also help the deaf and dumb to search the web or send mails more conveniently. It has two parts. One is Train part and second is sign detection part. Train part set by deep learning method using CNN network and make train dataset. The detection part takes sign from webcam and detect the Bengali numerical value by classifying from the train dataset.