Palm geometry biometrics: a score-based fusion approach

Abstract

In this paper we present an identification and authentication system based on hand geometry. First, we examine the performance of three different methods that are based on hand silhouette and on binarized hand images and then we investigate approaches for combining the feature vectors, identification-verification scores, and individual acceptance-rejections decisions taken by using each one of the proposed methods individually. The proposed system has been tested on the POLYBIO hand database which consists of 180 hand images from 45 individuals. The experiments show that fusion of feature vectors results in a slightly better performance in both identification and authentication tests while combination of scores and decisions leads to a significant improvement in authentication performance and minor improvement in identification.