Adaptive Hand Gesture Recognition System Using Machine Learning Approach

Abstract

One of the most important problem faced by our society is that deaf and hearing-impaired people are finding it hard to come up with the advanced moving technology. The sign language plays a vital role for these people as only through the sign language they can communicate with each other. But for other people it is very challenging to understand the sign language. But as we know, it is not feasible for everyone to learn sign language, need of a smart gesture recognition system for sign language is evolved. Sign language basically includes the various movements of palms, hands and other body parts. Gesture depicts an image of a specific body part. Gesture recognition is an advanced technology which is used to recognize the posture of body parts. Objective of this type of software is to provide medium between the human body movements with computer system. In gesture recognition system particularly hand gesture recognition provides a much effective and natural way for non-verbal data exchange when working with HCI system. This paper presents a real time vision approach for recognizing hand gesture based on machine learning using MATALB. Information gathering is a first step in our proposed approach where we will develop dynamic hand gesture dataset of sign language. In next step we will represent our original proposed framework to depict hand gesture recognition with the help of hand shape and its various motions descriptors. This second step will recognize the sign once it has been captured and is much challenging trade, especially in a continuous stream which a main focus of our work. In this paper we have presented an adaptive hand gesture recognition system using machine learning approach which is adaptable to multiple applications, thus making the gesture recognition systems to be application adaptive.