A New Method of Singular Points Accurate Localization for Fingerprint

To solve singular points (SPs) rapid and accurate localization problem, a novel algorithm of SPs rapid detection is presented. Based on preprocessed fingerprint image, orient field is calculated out and homogeneous-zones-divide. Candidate small areas are then selected by successive approximation and candidate singular points are chosen from candidate small areas by Poincare Index Algorithm (PIA). Finally, the real singular points are precisely located by the mean algorithm. To evaluate arithmetic performance, Experiments on FVC2004 DB2_B, DB3_B and DB4_B database were carried out. Compared with PIA, the experimental result show that singular points localization of our method is more precise, and the False Detecting Ratio (FDR) and the Missed Detecting Ratio (MDR) decrease by 5.86% and 6.8% respectively. At the same time, the speed of our algorithm is 7.6 times faster than PIA. This performance proved that our method present a feasible and effective way to detect and locate the SPs of fingerprint image.