Face recognition using wavelet packets decomposition and Hopfield neural network

Abstract

In this paper, we use edge information for face recognition. In order to get more details but less noise of an image, we decomposed the image with wavelet packets before the process of edge detection. After the edge detection, a logical data which only contains elements of 0 or 1 was introduced to the Hopfield neural network. The proposed algorithm is tested on ORL face database and the result is found to be perfect.