Fingerprint Identification Using Hierarchical Matching and Topological Structures

Fingerprint identification is one of the most popular and efficient biometric technique used for improving automatic personal identification. In this paper, we will present a new indexing based method, in the first step, on estimation of singular point considered as an important feature in the fingerprint using a directional file. In the second step, a hierarchical Delaunay triangulation was applied on the minutiae around the singular point extracted. Comparison of two fingerprints was calculated by introducing the barycenter notion so as to ensure the exact location of the similar triangles. 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.