Extracting information from point set for robust fingerprint authentication

Abstract

In biometric identification, a fingerprint is typically represented as a set ofminutiae which are 2-D points. We are interested in ways of extractinginformation from such a point set template for robust authentication. Thisthesis consists of two parts. In the first part, we show that analysis of anexisting method, Fingerprint vault[6], is inadequate. As such, the securityof such method may be too weak in practice. More specifically, we proposean attacker who achieves 2192.7 times speedup compared with a brute-forceattacker whom they used to measure the security of fingerprint vault. In thesecond part, we propose a new method to extract information from 2-D pointset by using feature lines and PCA transformation. We name this scheme asapproximate message authentication code for 2-D point set to produce MAClike binary code for robust fingerprint authentication. Given such a 128-bitlength AMAC code, to successfully launch a pre-image attack requires morethan 2^81 trials.