An orientation-based ridge descriptor for fingerprint image matching

Orientation field is a very useful global feature in Auto Fingerprint Identification Systems. It describes the basic shape, structure and direction of a fingerprint. Orientation-based fingerprint descriptor is independent of other minutiae; hence, it is robust to either spurious minutiae producing or genuine minutiae missing in minutiae detection algorithms. This paper proposes an orientation-based ridge descriptor, and uses it for fingerprint image matching. Restriction of slight nonlinear distortion is also considered for such a descriptor. We conduct experiments on FVC2000 DB1 to evaluate the performance of the proposed descriptor. A comparison is made between the proposed descriptor and a most widely used minutia descriptor. Experimental results show that our descriptor is more distinctive.