Detection of Singularities in Fingerprint Images Using Linear Phase Portraits

Abstract

abstract The performance of fingerprint recognition depends heavily on the reliable extraction of singularities. Common algorithms are based on a Poinc’are Index estimation. These algorithms are only robust when certain heuristics and rules are applied. In this chapter we present a model-based approach for the detection of singular points. The presented method exploits the geometric nature of linear differential equation systems. Our method is robust against noise in the input image and is able to detect singularities even if they are partly occluded. The algorithm proceeds by fitting linear phase portraits at each location of a sliding window and then analyses its parameters. Using a well-established mathematical background, our algorithm is able to decide if a singular point is existent. Furthermore, the parameters can be used to classify the type of the singular point into whorls, deltas and loops.