Singular Points Analysis in Fingerprints Based on Topological Structure and Orientation Field

Abstract

As an important feature of fingerprints, singular points (includingcores and deltas) not only represent the local ridge pattern characteristics, but also determine the topological structure (i.e. fingerprint type). In this paper, we have performed analysis for singular points in two aspects. (1) Based on the topology theory in 2D manifold, we deduced the relationship between cores and deltas in fingerprints. Specifically we proved that every completely captured fingerprint should have the same number of cores and deltas. We also proposed a flexible method to compute the Poincare Index for singular points. (2) We proposed a novel algorithm for singular point detection using global orientation field. After the initial detection with the widely-used Poincare Index method, the optimal singular points are selected to minimize the difference between the original orientation field and the model-based orientation field reconstructed from the singular points. The core-delta relation is used as a global constraint for final decision. Experimental results showed that our algorithm is rather accurate and robust.KeywordsSingular PointTopological StructureMachine IntelligenceGlobal ConstraintOrientation FieldThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.