A Novel Palmprint Segmentation and Recognition Algorithm

Abstract
According to palmprint region segmentation problem, a novel five-point segmentation algorithm is proposed in this paper. Firstly, the maximum inscribed circle should be located in palm image by using fast method based on affine transformation, and three junctions of finger and finger can be located in polar coordinates, and then the two tangential points can be gotten by using the method of polar coordinates transformation. Finally, palmprint segmentation can be realized based on these points. The experiment results show that the algorithm is accurate and effective. After achieving the effective region, palmprint pattern's classification experiments are done, and the results show that the recognition algorithm is effective.

This publication has 7 references indexed in Scilit: