Palmprint recognition by a two-phase test sample sparse representation

The development of accurate and robust palmprint recognition algorithm is a critical issue in automatic palmprint recognition system. In this paper, we propose a palmprint recognition method based on a two-phase test sample sparse representation. In the first phase, a test sample is represented as a linear combination of all the training samples and m "nearest neighbors" are selected based on the representation ability. In the second phase, the test sample is represented as a linear combination of the determined m nearest neighbors and the representation result is used for classification. Experimental results on PolyU database show the effectiveness of the proposed method in terms of recognition rate.