Fingerprint Matching Algorithm

In recent years, automatic fingerprint identification technique (AFIT) has been a research focus in the area of science. Many researchers have done much work on fingerprint identification. But some technical difficulties always baffle the development of fingerprint identification. The reliable, strong fingerprint-matching algorithm has important meaning to improving the speed and the performance of system. Meanwhile, fingerprint sensor diversification has put forward higher requirement to AFIT, and interoperability of many different kinds of fingerprint sensors has important meaning to broad applications of AFIT. Therefore, in this paper, we deeply analyze and investigate the key algorithm of automatic fingerprint identification system----fingerprint-matching algorithm.We investigate two algorithms of fingerprint matching based on OBMD (orientation-based minutia descriptor) — global matching algorithm and overlap matching algorithm after fingerprint image alignment. For two algorithms, computer match score and analyze match results, respectively. OBMD is a new representation of minutiae. It consists of not only location and direction of minutiae, but also orientation of broad area around the minutiae. In addition, it has a useful property that it is independent with respect to other minutia detected in image. The algorithm based on OBMD belongs to minutiae-based algorithm. On the other hand, it imposes minutiae and fingerprint orientation field, so it also belongs to hybrid feature matching algorithm.In different idea, we think of a kind of matching using relation among minutiae in a local area. In allusion to the limitation of minutiae-based algorithm and effect of image rotation and translation on fingerprint matching, we propose an fingerprint matching algorithm based on DT (Delaunay Triangulation) net, which is stable, and has rotation and translation invariability. It also belongs to minutiae-based algorithm. On the principle of Delaunay, triangulation uses a set of minutiae to form a DT net. There are many local structures in net, for instance, triangles and quadrangles. In a triangle, length of edges and inclination between edges are also used in matching process. But, every...