A Comprehensive Study on Feature Extraction and Classification Techniques for Sign Language Recognition

Abstract

Sign language is the single means by using which the deaf community can interact with individuals living across the world. However, a small population of normal people can comprehend sign language adequately. As a result, a good Sign Language recognition system is needed to assist bridge the gap between normal people and persons with physical disabilities. This study presents a comprehensive overview of state-of-the-art feature extraction and classification techniques utilized in recent research on the recognition of sign language and hand gestures. Overall, the goal of this research is to give readers a thorough overview of the field of automated gesture and sign language recognition as well as to inspire more research work in the respective domain. The number of existing papers is reviewed based on used techniques and acquired accuracy. As a result, it is analyzed that the most reliable image processing and feature extraction techniques are vision-based techniques, whereas, for classification, the most efficient techniques are machine learning techniques.