Fingerprint matching using local alignment based on multiple pairs of reference minutiae

摘要

A novel minutiae-based method is proposed to matchndeformed fingerprints. We combine minutiae structures and descrip-ntors to obtain multiple pairs of reference minutiae and globally alignntwo sets of minutiae to get a common overlapping region based onnthese pairs of reference minutiae. At the matching stage, our pro-nposed approach aligns local areas in two fingerprints with new ref-nerence minutiae pairs that are selected from matched minutiaenpairs. After registration of the fingerprints according to the local cor-nrespondence, the number of matched minutiae can be counted us-ning tight thresholds. Experimental results confirm that the proposednalgorithm is effective for fingerprint matching with nonlinearndistortions.