An algorithm for fingerprint basic minutiae feature matching

Abstract

Minutiae matching is a very important procedure in a fingerprint recognition system. In comparison with usual minutiae matching approach, this paper proposes a minutiae matching approach only with basic minutiae feature, i.e. minutiae position and minutiae direction. The approach includes three procedures, i.e. fingerprint alignment, minutiae matching, and similarity level computation. Experimental results show that the EER (Equal Error Rate) of proposed approach is 0.075 on FVC2004 DB1. The proposed approach is with only basic minutiae, therefore, it achieves a good trade-off between EER and template size. Therefore it is suitable for some specific situations that space is scarce resource.