Identity Verification Through Dorsal Hand Vein Texture Based on NSCT Coefficients

Abstract

This paper presents a novel method for person recognition using dorsal hand vein texture image based on phase response information of NonSubsampled Contourlet Transform (NSCT). In Our approach, pre-processing phase is applied on the image contrast in order to produce a better quality of dorsal hand vein image, then, we localize the ROI and we analyze its texture by NSCT. Next, we have proposed a novel encoding method based on extracting statistical descriptors from local region of the dorsal hand vein to create a code of 512 bytes. Finally, we have calculated the modified Hamming distance between templates to find out the similarity between two dorsal hand vein filtered images. The method evaluation is completed on GPDSvenasCCD database and experimental results are compared with other methods. The experimental results illustrate the effectiveness of this coding in verification mode of biometric palm vein: 0.10% of equal error rate. Therefore, the coding process achieve more satisfactory and convincing results than performed by well-known approaches and is proved the robustness of NSCT method to extract discriminative features of dorsal hand veins texture.