Hand gesture recognition system with real-time palm tracking

Abstract

Over the years Vision based real time gesture recognition system has witnessed an exponential growth because of its manifoldness applications, ranging from sign language to virtual reality and its ability to interact with system efficiently through HCI. In this paper, we have proposed a hand gesture recognition system for American Sign Language recognition using important features of hand such as fingertips, palm center etc. The system is capable of recognizing hand gestures even when the fore-arm is involved and it can tolerate a certain rotation of palm and fore-arm. We have implemented principal component analysis to remove ambiguity between two similar types of gestures and given emphasis to detect movement epenthesis by means optical flow technique.