On separating overlapped fingerprints

Abstract

Fingerprint images generally either contain only a single fingerprint or a set of non-overlapped fingerprints (e.g., slap fingerprints). However, there are situations where more than one fingerprint overlap on each other. Such situations are frequently encountered when latent fingerprints are lifted from crime scenes or residue fingerprints are left on fingerprint sensors. Overlapped fingerprints constitute a serious challenge to existing fingerprint recognition techniques, since these techniques are designed under the assumption that fingerprints have been properly segmented. In this paper, a novel algorithm is proposed to separate overlapped fingerprints into component or individual fingerprints. We first use local Fourier transform to estimate an initial overlapped orientation field, which contains at most two candidate orientations at each location. Then relaxation labeling technique is employed to label each candidate orientation as one of two classes. Based on the labeling result, we separate the initial overlapped orientation field into two orientation fields. Finally, the two fingerprints are obtained by enhancing the overlapped fingerprint using Gabor filters tuned to these two component separated orientation fields, respectively. Experimental results indicate that the algorithm leads to a good separation of overlapped fingerprints.