Application of improved BOMP algorithm in face recognition

When the group sparse representation is used to face recognition, the same samples take participate in representation the test sample at the same time. The original method ignored the correlation between the samples. To solve this problem, an improved block orthogonal matching pursuit algorithm is presented. The proposed algorithm uses the coherent coefficient of the samples as a parameter, setting the proper threshold value to select sample discrimination. Therefore, the reconstruction of the algorithm is optimized. Experiments on the Yale B database show that the recognition rate of improved algorithm is higher than the original one. The experiment results verify the validity of the proposed algorithm.