Derivative Method for Hand Palm Texture Biometric Verification

Abstract

In this work, a novel, simple and very robust biometric verification system through the use of the texture of the hand palm is proposed. It attempts to make the performance of existing palm-print systems better. First of all, the hand palm image with scale, rotation and translation invariance is isolated from the hand image recorded. Then, the "derivative method" presented in this paper, is used to extract the texture features from gray-scale images. It consists of a differentiation and binarization process. 1090 hand images of 109 people with 10 samples each one have been acquired by means of a commercial scanner with 150 dpi resolution. Support Vector Machine (SVM) is the main classifier used as verifier, in closed mode and open mode. An EER=0.30% and an EER=0.032% shown the final results of our system when it works in open and closed mode respectively.