A Technique for Hand Gesture Recognition on Real Time Basis

Abstract

Sign gesture is a non-verbal visual language, different from the spoken language in terms of medium of communication, but serves the same function for hearing & speech impaired community. Gesture Recognition, and more specifically hand gesture recognition, is one of the typical methods used in sign language for non-verbal communication. It is often very difficult for the hearing & speech impaired community to communicate their ideas and creativity to the normal humans. This paper focuses on discussing different methods to identify the gesture. Method for hand segmentation is discussed in terms of the different approaches to sub-components of the identifying the gesture. The judgement parameters are accuracy in real time performance, processing time, processor utilization, etc.

Real Time Conversion of Hand Gestures to Speech using Vision Based Technique

“…The tips are correlated with the centre of the palm to identify the respective gesture. Skin color is mostly used for hand recognition various color models suggested in [13] are cbcr color space, RGB color Space, HS (Hue , Saturation value)color space, Normalized RGB & HSV .Trong-Nguyen, Nguyen [3] used skin color, Md Iqbal Quraishi, Krishna Gopal Dhal, J Paul Choudhury [4] used YCbCr color, S. Bhowmick, S. Kumar and A. Kumar [10] have a hybrid HSV+YCbCr colour model for detecting the hand, [3]filling the gaps and mapping to stored dataset. [4] used neural networks for identifying gesture.…”

“…Sign Language is the non-verbal way of communication usually used by people having hearing impairment. In sign language, gestures are considered to be any specific patterns or movements of the hands, face or body to make certain expressions [10].Gestures can be expressed with the help of facial expressions, limb movements or any meaningful bodily state [8]. Gesture can be static or dynamic.…”