A Bounding box Gabor filter-based Minutiae extraction scheme for fingerprint Recognition system

Abstract

A complete bounding box Gabor filter based minutiae extraction scheme for fingerprint recognition system is presented. The proposed method uses improving alternatives for the image enhancement process, leading consequently to an increase of the reliability in the minutiae extraction task. In the first stages, image normalization, median filtering and orientation field of the fingerprint are calculated. The local orientation of the ridges will serve as parameter for the next processing stages. Details for the morphological filtering used to ridge extraction and back ground noise elimination are described. Conclusions in terms of GAR (Genuine Acceptance ratio), FAR (False acceptance ratio) and FRR (False rejection ratio) of this method are compared with Gabor filter based approach.