Non-Alignment Fingerprint Matching Based on Local and Global Information

This paper proposes a novel triangle descriptor taking advantage of global orientation and local triangle structure information for fingerprint representation. The representation allows the derivation of a fixed-size feature vector with which we can reliably achieve corresponding triangles that is invariant to the rotation and translation of the fingerprint. With these corresponding triangles as seeds, we develop a non-alignment fingerprint matching scheme based on triangular framework. Experimental results show that the proposed method performs better than other approaches