Real time gesture recognition and processing to control television set by hand beacon

The goal of this research is to develop a system for real time gesture recognition and classify the sign and send to TV. A continuous stream of hand signs images is recorded and has to be processed immediately. The camera that records these signs provides pictures in real-time conditions to a PC that has to classify the signs. Such systems should offer the possibility to command computers just using hand signs. To simplify the task the research tried to find a robust solution for recognizing signs correctly as well as make the computation faster and interface with television to supersede the remote controller. The system function is divided into four major parts. Firstly, from the hand gesture some features have been extracted. To calculate height and area of the gesture, centroid and distance of that centroid from the origin are the extracted feature. Secondly, an input array is formed using these features and converted it into formatted feature matrix. For this the decimal input matrix has been converted into binary matrix by using a ranging method. The test image is collected from the camera of the computer in real time environment. Then, this binary matrix is used as the input of a feed-forward back-propagation neural network and tries to find the decimal output from it and convert it into binary. Finally, that corresponding character of channel is sent through serial port to the television via web cam.