Sign Language Recognition using Machine Learning

Abstract

Deaf and dumb people communicate with others and within their own groups by using sign language. Beginning with the acquisition of sign gestures, computer recognition of sign language continues until text or speech is produced. There are two types of sign gestures: static and dynamic. Both gesture recognition systems, though static gesture recognition is easier to use than dynamic gesture recognition, are crucial to the human race. In this survey, the steps for sign language recognition are detailed. Examined are the data collection, preprocessing, transformation, feature extraction, classification, and outcomes. There were also some recommendations for furthering this field of study.