Palm vein recognition by using modified of local binary pattern (LBP) for extraction feature

Abstract

Palm vein recognition is developing biometric identification technology. It can be used in physical security and information security for selective control of access to a place or resource. A palm vein recognition has been gaining research interest from last few years because it use physiological intrinsic that uniqueness, stability, not easily spoofed and damaged and have live body identification. There are consists of the following steps: Image acquisition from the database and Pre-Processing, Finding of Region of interest, Extraction of Palm Vein pattern Features and Matching. Prior to the palm vein recognition, vein extraction is generally required for a better recognition. In this paper we propose a vein extraction method modified of the Local Binary Pattern (LBP) combining with Probabilistic Neural Network (PNN) for matching. The aim of the proposed system is to improve the accuracy of palm vein recognition. Simulation result show that the proposed method has a higher recognition rate for palm vein recognition comparing to the other basic Local Binary Pattern.