A minutiae level fusion for AFIS systems

Minutiae correspond essentially to the termination and bifurcation of fingerprint patterns. Since the quality of fingerprint images is often low, extraction algorithms produce a large number of false alarms. A complete minutiae extraction scheme for automatic fingerprint recognition systems is presented. A great deal of work has been dedicated to fingerprint enhancement. We propose a new fusion approach to combine different extraction algorithms. The proposed method, as confirmed by simulations, allows a good overall performance to be achieved suitable for identification forensic applications.