Fingerprint Matching Incorporating Ridge Features

Abstract

This paper introduces novel fingerprint matching
based on ridge features and conventional minutiae features to increase non-linear deformation in fingerprints. As an extension we propose to use bifurcations along with ridge patterns. The ridges along with bifurcations are considered as minutiae. Our proposal also finds either of these three points even in distorted and noisy fingerprints. This kind of multiple feature have some advantage in that they can represent authentication using fuzzy optimization provides an incomparable fingerprint matching. The false acceptance
minimization is the most advantageous achievement of this proposal. We use crossing number algorithm to detect minutiae points. To reduce false minutiae we propose fuzzy rules. Experiments were conducted for the FVC2002 and FVC2004 databases to compare the proposed method with the conventional minutiae-based method. This type of feature extraction allows a best fingerprint matching.