3D finger biometrics

Abstract

This paper investigates the use of the back surface of the hand, specifically the fingers, as a biometric feature. We focus our efforts on the index, middle and ring fingers. We use segmented 3D range images of the back of the hand. At each pixel lying on one of the these fingers, the minimum and maximum curvature values are calculated and then used to compute the shape index, resulting in a shape index image of each of the three fingers. The shape index images are then compared to determine the similarity between two images. We use data sets obtained over time to analyze the stability of this feature, and examine the performance of each finger as a separate biometric feature along with the performance obtained by combining them. Our approach yields good results indicating this approach should be further researched. © Springer-Verlag Berlin Heidelberg 2004.