A Novel Topology-Based Matching Algorithm for Fingerprint Recognition in the Presence of Elastic Distortions

摘要

Fingerprint matching is a crucial step in fingerprint identification. In this paper, we propose a novel fingerprint alignment and matching scheme based on the geometric structure of minutiae set. The proposed scheme is transformation independent and is based on the Delaunay triangulation based technique. We also propose a novel application of RBF method, typically used for medical image matching, in order to deal with elastic deformations of the fingerprint. Finally, we develop an efficient global matching scheme based on the comparisons of minutiae sets and singular points to increase the validity of the matching results. Using the same amount of information on minutiae set as traditional minutiae based algorithms we achieve a faster performance and higher accuracy rates using our proposed algorithm.