Hand Gesture Recognition Using Backpropogation Algorithm Based on Neural Network

Abstract

The proposed system is design for physically challenged people to communicate among common people without an intermediate human translator. The system based on hand gesture recognition using neural network. We used back propogation algorithm for the recognition of image. Some of the systems are used for the purpose of communication are costly and bulky that the common man can not afford but system we are going to used is affordable so it is possible to minimize the distance between hearing and speech impaired people with normal human being. The image is captured with the help of inbuilt camera of laptop comparing with the existing database using matlab and further it will process in neural network accordingly gives the output in the form of text along with the accuracy in percentage. Firstly the prepocessing steps are completed. The steps going to perfom are image aquisiton, image processing, feature extraction, gesture identification and finally output translated in text. We calculate the centroid of the hand image called as vectorization which will futher train our neural network and after processing we get output.
DOI: 10.17762/ijritcc2321-8169.150760