Directional Gaussian Derivative Filter Based Palmprint Authentication

This paper presents a novel approach of palmprint authentication based on the directional gaussian derivative filters. In this approach, the palmprint image is firstly filtered using some directional Gaussian derivative filters in different directions. Then the orientation and the phase of each pixel of the palmprint image are extracted to form a feature. In the matching phase, the Hamming distance is employed to measure the similarity of different palmprint. The experimental results show that the proposed approach can effectively discriminate between palmprints and the EER is about 0.139%.