Fingerprint identification using hierarchical matching and topological structures

Abstract

Fingerprint identification is one of the most popular and efficient biometric techniques used for improving automatic personal identification. In this paper, we will present a new indexing method, based on estimation of singular point considered as an important feature in the fingerprint by using the directional file. On the other hand, a hierarchical Delaunay triangulation is applied on the minutiae around the extracted singular point. Two fingerprints calculated by introducing the barycenter notion to ensure the exact location of the similar triangles is compared. We have performed extensive experiments and comparisons to demonstrate the effectiveness of the proposed approach using a challenging public database (i.e., FVC2000), which contains small area and low-quality fingerprints.