Hand gesture recognition using a convolutional neural network for arabic sign language

Abstract

Communication is one of the most basic human needs, but for between deaf and dumb people and normal people, communication in their daily lives is a challenge, due to the lack of reliable and easy-to-use assistive devices and skilled sign language interpreters. The need to integrate hard-of-hearing Arab individuals into their societies has recently received greater attention than many public and private institutions. Accordingly, in the fields of artificial intelligence(AI) and machine learning(ML), automating sign language recognition has become a critical technology, this paper presents a proposed system for static hand gestures recognition for the Arabic sign language (ArSL) using a Convolutional Neural Network(CNN) with accuracy 97.42 %.