Moroccan sign language recognition based on machine learning

Abstract

More than 5% of the world's population (466 million people) suffer from a disabling hearing loss: 4 million are children. People with hearing loss usually communicate through spoken language and can benefit from assistive devices such as cochlear implants. However, deaf people have profound hearing loss and use sign language to communicate with others, which involves little or no hearing. To facilitate communication between deaf people and normal people who do not know sign language, we have proposed in this paper a system that allows textual transcription of sign language. The developed system will be able, in a first step, to recognize the sign language alphabet using machine learning and image processing. Simulation results have shown the efficiency of the developed model.