Abstract
In this paper, a collaborative representation based two-phase face recognition method is proposed. In the first phase, the test sample is represented by a linear combination of all the training samples, and then the sum of contributions of each class is calculated. As a consequently, we use the sum of contributions to determine k classes of training sample that have the maximum sum of contributions for the test sample. In the second phase, the test sample is also represented by a linear combination of the k classes of training sample. As a result, we use the representation result of each class to reconstruct the collaborative image of the test sample. Moreover, the face classification is performed by using the similarity measures including structure similarity index measure (SSIM), root mean square (RMS), and similarity assessment value (SAV). The experimental results show that our method outperforms the two-phase test sample representation method (TPTSR).