Improved face recognition based on the fusion of PCA feature extraction and sparse representation

Sparse representation is a hot research topic in the field of biometrics in recent years. Even though the image has different expression and illumination, as well as occlusion, most of the algorithms still have good recognition effect. However, when the image contains illumination and facial expression changes, sparse classification method does not have good robustness. In this paper, we present a method that combination the PCA feature extraction method and sparse representation, which is a simple and effective face recognition algorithm. Based on the sparse concentration index, we propose the index threshold. When the concentration index is lower than the threshold value, we need to choose five training samples which corresponding to the minimum residuals, then construct a new training sample dictionary, and use the reconstructed dictionary to classify the test samples. This method strengthening the verification of abnormal test samples and optimizes the classification strategy. Comparing with the traditional SRC method, the simulation results show that the improved method is better than the SRC classification.