Empirical tests on BioHashing

Abstract

Given the recent explosion of interest in human authentication, verification based on tokenised pseudo-random numbers and user-specific biometric feature (BioHashing) has received much attention. The purpose of our paper is to investigate the pros and the cons of BioHashing. In particular we study: how to combine BioHashing in a multi-modal fusion approach based on the integration of face and fingerprint data; how to combine BioHashing and user specific Iris features. The results of this paper confirm that a multi-modal biometric can overcome some of the limitations of a single biometric resulting in a substantial performance improvement. Moreover, we show that a multi-matcher system based on the fusion between "BioHashed" Iris features and a "standard" Iris matcher permits to obtain good performance when an "impostor" steals the pseudo-random numbers, and a near zero Equal Error Rate when nobody steals the pseudo-random numbers. © 2006 Elsevier B.V. All rights reserved.