A Palm-vein recognition algorithm based on LPP and HM-LBP

Abstract

In this paper, an algorithm based on locality preserving projection (LPP) for palm-vein was proposed. The feature of palm-vein was extracted by Hierarchical multiscale local binary pattern(HM-LBP), and then the dimension was reduced by LPP. Lastly, the palm-vein was classified by normalized correlation classifier. The algorithm was tested on the near- infrared database of the multi-spectral palm database of Hong Kong polytechnic university. The experiment results show that the algorithm can achieve ideal recognition accuracy of 99.92% and the speediness is able to fit for the real-time palm-vein recognition system.