Fusion of multiple features for palmprint authentication

Abstract

A palmprint image contains abundant features, such as the ridges, lines and textures, etc. Different features reflect the different characteristic of the palmprint. Fusion of multiple palmprint features may enhance the performance of a palmprint authentication system. In this paper, we investigate the fusion of two novel palmprint representations: orientationCode and diffCode. OrientationCode are defined using four directional templates and diffCode is computed using differential operation. Then these two codes are fused to measure the similarity of palmprints. The experimental results show that the proposed approach can get a very high accuracy.