A minutia-based partial fingerprint recognition system

Abstract

Matching incomplete or partial fingerprints continues to be an important challenge today, despite the advances made in fingerprint identification techniques. While the introduction of compact silicon chip-based sensors that capture only part of the fingerprint has made this problem important from a commercial perspective, there is also considerable interest in processing partial and latent fingerprints obtained at crime scenes. When the partial print does not include structures such as core and delta, common matching methods based on alignment of singular structures fail. We present an approach that uses localized secondary features derived from relative minutiae information. A flow network-based matching technique is introduced to obtain one-to-one correspondence of secondary features. Our method balances the tradeoffs between maximizing the number of matches and minimizing total feature distance between query and reference fingerprints. A two-hidden-layer fully connected neural network is trained to generate the final similarity score based on minutiae matched in the overlapping areas. Since the minutia-based fingerprint representation is an ANSI-NIST standard [American National Standards Institute, New York, 1993], our approach has the advantage of being directly applicable to existing databases. We present results of testing on FVC2002's DB1 and DB2 databases.

Introduction

Fingerprint matching based on minutia features is a well researched problem. During the last four decades, various algorithms have been proposed to match two minutia templates of fingerprints. Most of these algorithms assume that the two templates are approximately of the same size. This hypothesis is no longer valid. Miniaturization of fingerprint sensors has led to small sensing areas usually varying from 1” $\times$ 1” to 0.42” $\times$ 0.42”. However, fingerprint scanners with a sensing area smaller than 0.5” $\times$ 0.7”, which is considered to be the average fingerprint size [2], can only capture partial fingerprints.

Matching small (partial) fingerprints to full pre-enrolled images in the database has several problems: (i) the number of minutia points available in such prints is few, thus reducing its discriminating power; (ii) loss of singular points (core and delta) is likely and therefore, a robust algorithm independent of these singularities is required; and (iii) uncontrolled impression environments result in unspecified orientations of partial fingerprints, and distortions like elasticity and humidity are introduced due to characteristics of the human skin.

A minutiae-based fingerprint matching system usually returns the number of matched minutiae on both query and reference fingerprints and uses it to generate similarity scores. Generally, more matched minutiae yield higher similarity scores. That is when the number of minutiae on both fingerprints is large we can confidently distinguish the genuine and imposter fingerprint using the number of matched minutiae. According to forensic guidelines, when two fingerprints have a minimum of 12 matched minutiae they are considered to have come from the same finger [3]. However, it is not reasonable to use an absolute number of matched minutiae alone in case of partial fingerprints. We must also consider the overlapped areas on both prints and the total distance between all the matched minutiae to obtain a similarity score.

In this paper, we discuss the various issues involved in the matching of such partial fingerprints. In Section 2, we outline our approach to partial fingerprint matching. A localized secondary feature matching is described. The secondary features are derived from minutiae features. It does not depend on global ridge structures (e.g. core and delta) making it suitable for matching partial fingerprints. The notion of dynamic tolerance area and generation of reliable similarity score is discussed. In Section 3, we briefly review the minimum-cost flow (MCF) problem in the context of fingerprint matching. This technique is applied to both secondary feature matching and the brute-force matching derived from minutiae features. Experimental results are presented in Section 4.