Abstract—The aim of this paper is to provide with a real time application based on hand gesture recognition. Virtual touch is a concept whereby any normal surface with no internal circuitry and hardware for touch sensitivity, is converted into a touch sensitive surface by using image processing techniques. Using such a technique the paper presents the transformation of a plain white wall into a large touch wall. The virtual touch wall aims at building a gesture sensitive environment for users, to provide them new ways of interaction in their work atmosphere and in their recreational space. The hardware requirements of the system are kept on the minimum , limiting only to a simple USB webcam and a PC .Other alternatives to the PC have also been presented , such as the Beagle Board , which is a mini version of a PC and very inexpensive .The approach is based on simple and fast motion detection trying to eliminate unnecessary regions of interest and a recognition algorithm based on the hand contours , their convexities and a further comparison of hand shapes based on HU moment matching which works on image contours .The paper also presents a way of improving the distorted hand contours by distinguishing between external contours and holes and explicitly filling up holes .The use of a predefined set of contours for each gesture helps in better recognition .A simple state machine is implemented to further improve reliability .A total of 5 gestures have been implemented and tested on the PC and the Beagle Board and the performance compared. The Linux XLIB library and the X display functions have been utilized to control mouse cursor actions and other display properties of the wall.
Index Terms—Gesture Recognition, HU Moments, Hand Tracking, Contours, Xlib, Beagle Board, virtual gesture wall
P. N. V. S. Gowtham is with the Department of Electronics and Communications, International Institute of Information Technology (e-mail: email@example.com).
Cite: P. N. V. S. Gowtham, "A Hand Gesture Recognition Based Virtual Touch Wall," International Journal of Information and Education Technology vol. 2, no. 1, pp. 36-42, 2012.