Joint palmprint and palmvein verification by Dual Competitive Coding

Abstract

Palmprint and palmvein based personal verification has attracted lots of attention recently in the biometrics society. As both the two biometric features can be captured from the palm, it is intuitive and necessary to develop a multimodal system by using the two features for better recognition accuracy. In this paper, a personal verification system which fuses palmprint and palmvein patterns is proposed. A specific device is designed to have a near-infra-red (NIR) camera and an NIR illumination source, so that the palmprint and palmvein information can be captured in a single image simultaneously. In order to represent the two types of biometric features effectively and efficiently, a novel coding scheme, namely Dual Competitive Coding, is proposed. Experimental results on a large database show that the proposed scheme could achieve high recognition accuracy in high matching speed.