Dorsal hand-vein images recognition system based on grey level co-occurrence matrix and Tamura features

Abstract

Biometrics is the important area of distinguishing people using their behavioural characteristics. Until now, researcher and exporter increasing interest with vein pattern biometrics. A vein pattern is a massive link of blood vessels under a person's skin. Similar to fingerprints, in scientific sense, the shape of vascular patterns in the same part of the body has proved distinct from each other. The objective of this paper is to analyse vein images and to design and implement dorsal hand-vein recognition system that has the ability to segment vein and recognise each person based on his vein requires the presence of the human operator. The experimental results indicate that the MDC classifier achieves accuracy of 92% in the case of wavelet transform, and GLCM and Tamura features.