Automatic system for Arabic Sign Language Recognition and translation to spoken One
International Journal of Advanced Trends in Computer Science and Engineering · 2020
The deaf-mutes person suffers from inability to speak as well as inability to hear others.This disability creates a barrier that prevents the disabled from integrating with society.Therefore, the owners of this disability attempt to break this barrier by using many methods to try to communicate with others.The most famous of these methods is the sign language which is relying on the body language.The disabled person uses facial expressions and hand gestures to express his needs.In this project, we will try to build a computerized system depending on the depth-measuring cameras and computer vision techniques to capturing and segmenting the pictures of the facial expressions and hand gestures.The segmented gestures are classified and stored in an electronic library after they have been recognized and linked with their corresponding spoken and written Arabic words.The computerized system will be learned with gestures and its corresponding spoken and written Arabic words.A gesture dataset of 40-segmented words that frequently used in the life by the hearing-impaired person was utilized for testing the system.We get in our consideration that each word may has a different occlusion state.The experimental results illustrated that the proposed system has a recognition accuracy over 90%