A Novel Verification Criterion for Distortion-Free Fingerprints

摘要

An important aspect of fingerprint verification systems is the method used to quantify the similarity between two fingerprints. This involves two key components: choosing fingerprint features that will be used for comparison and selecting a match score function to calculate the degree of correspondence. The choice of features and a match score function can have a significant impact on the performance of a system. This paper presents a novel fingerprint verification criterion based on tabulating ridge intersections between distortion free fingerprints. Several alternative matching criteria have been implemented, and their performance is compared using a publicly available FVC2002 dataset. The novel ridge based approach proves to be highly discriminative, and a strong result is obtained by a hybrid system using a combination of minutiae and ridge based features.