Face Recognition Based on Log-Gabor Feature and Collaborative Representation

Abstract

In this paper, a face recognition method based on Log-Gabor feature and collaborative representation is proposed, which aims at the problem of Collaborative Representation which has low robustness in illumination change, expression change and occlusion. Firstly, Log-Gabor features of the images are extracted, then the face images are divided into blocks, finally the Collaborative Representation is used to compute the category of image. The algorithm was evaluated using three well-known face databases of Extended Yale B, AR and ORL, the experiment results show that this algorithm has better recognition rate.