Principal axis and crease detection for slap fingerprint segmentation

In slap fingerprint segmentation, crease is the most difficult edge to correctly detect. In this paper, we present a novel yet simple and accurate algorithm for the principal axis and crease detection. Firstly, the principal axis of each foreground region is detected using the minimal rotational inertia; Secondly, the crease detection is done based on cost function minimization. This algorithm has been incorporated in a slap fingerprint segmentation scheme, previously developed by the authors, producing successful results.