Abstract:
People with disabilities have long been ignored. With the advancement of recent technologies, so many tools and software are designed for disabled people to improve their lives. In this research, the Arabic Sign Language (ArSL) recognition system is developed using the proposed architecture of the Deep Convolutional Neural Network (CNN). The aim is to help people with hearing problems to communicate with normal people. The proposed system recognizes the signs of the Arabic alphabet based on real-time user input. The Deep CNN architectures were trained and tested using a database of more than 50000 Arabic sign images collected from random participants of different age groups. Several experiments are performed with changing CNN architectural design parameters in order to get the best recognition rates. The experimental results show that the proposed Deep CNN architecture achieves an excellent accuracy of 97.6%, which is higher than the accuracy achieved by similar other studies.