Fingerprint Matching Algorithm Based on Similar Vector Triangle

Fingerprint matching is the key component of Automated Fingerprint Identification System (AFIS). Not only the online and real-time fingerprint identification system, but also the fingerprint identity key and the challenge-response authentication based on biological feature, acquires fingerprint matching algorithm based on minutiae with high accuracy and robustness. This paper introduces a novel fingerprint matching algorithm based on minutiae, which doesn´t rely on fingerprint global feature, such as core points and ridge structure. The algorithm uses a method of similar vector triangle to solve the key problem of point pattern matching algorithm reference point positioning. It can keep off a common situation happening in the traditional reference point positioning based on congruent or approximate congruent triangle which assumes the degree of deformation. Experimental results show that this algorithm can be invariant to translation, rotation and deformation. And it proves to be robust for fingerprint matching.