Fingerprint analysis and singular point detection

Correctly locating singular points (core and delta points) is crucial for most fingerprint classification and recognition applications. In this paper, we propose an algorithm to compute pixel direction and in return create essential primitive features called fault lines. By analyzing direction sequence of fault lines, we are able to provide a computational definition of singular points and distinguish different types of singular points. We also present a shrinking and expanding algorithm (SEA) based on a scale-pyramid model to extract singular points within an area as small as 2 × 2 pixels from fingerprint images. Our algorithm is rotation insensitive and can be applied to all types of fingerprints. Fingerprint images from the FVC2004 database are used for an experimental test, and the accuracy rate of the algorithm on identifying singular points is 92.2% (97.6% for core and 83% for delta points).