Biometric security: Palm vein recognition using Lbp and sift

Abstract

Palm vein recognition is an unique vascular recognition technique used for identification of individual. Palm vein recognition offer high degrees of security as vein are inside the skin so therefore difficult to forge. Palm vein pattern are unique for any individual so this can also be used to grant access to the user instead of using passwords, identification cards etc. In this paper we have used the combination of pixel wise Local Binary Pattern (LBP) and Scale Invariant Feature Transform (SIFT) technique to extract the palm vein features to improve the accuracy. In this paper the authors have presented a systematic comparison of some of previous palm vein recognition techniques with a novel technique proposed based on PolyU database. Evaluation of improvement in performance for recognition and verification process has been carried out and thereafter an elaborate analysis has been done on the effect of the size of enrolment. Simulation results depict an improvement of recognition rate and false acceptance rate. Implementation of the proposed method has been carried out using Image Processing Toolbox under MATLAB software.