Comparative analysis of global hand appearance-based person recognition

bstract. We provide a survey of hand biometric techniques in the iterature and incorporate several novel results of hand-based peronal identification and verification. We compare several feature ets in the shape-only and shape-plus-texture categories, emphaizing the relevance of a proper hand normalization scheme in the uccess of any biometric scheme. The preference of the left and ight hands or of ambidextrous access control is explored. Since the usiness case of a biometric device partly hinges on the longevity of ts features and the generalization ability of its database, we have ested our scheme with time-lapse data as well as with subjects that ere unseen during the training stage. Our experiments were conucted on a hand database that is an order of magnitude larger than ny existing one in the literature. © 2008 SPIE and IS&T. DOI: 10.1117/1.2890986