Fingerprint Matching Using Ridge Features and Hamming Algorithm for Distorted Fingerprints

This paper proposes a new fingerprint matching system which uses both ridge features and conventional minutiae features. This system may increase the recognition performance of fingerprint matching system against skin distortions. This system suggests the extraction of ridge features like ridge count, ridge length, ridge curvature direction and ridge type. These ridge features are less affected by skin distortions. To extract the ridge features we use the ridge based coordinate system in the skeletonized image. With the proposed ridge features and conventional minutiae features we propose a matching procedure using the hamming algorithm. In the matching process we compute the Euclidian distance between the input fingerprint finger code and the finger code of the template. FVC 2002 was used as the sample database of fingerprints. Thus we conclude that the matching process using the proposed features and hamming algorithm may increase the recognition performance of the fingerprint recognition systems against distorted fingerprints.