Please use this identifier to cite or link to this item:
Title: Singular points analysis in fingerprints based on topological structure and orientation field
Authors: Zhou, J
Gu, J
Zhang, D 
Issue Date: 2007
Publisher: Springer
Source: Lecture notes in computer science (including subseries Lecture notes in artificial intelligence and lecture notes in bioinformatics), 2007, v. 4642 LNCS, p. 261-270 How to cite?
Journal: Lecture notes in computer science (including subseries Lecture notes in artificial intelligence and lecture notes in bioinformatics) 
Abstract: As an important feature of fingerprints, singular points (including-cores 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.
Description: 2007 International Conference on Advances in Biometrics, ICB 2007, Seoul, 27-29 August 2007
ISBN: 9783540745488
ISSN: 0302-9743
EISSN: 1611-3349
Appears in Collections:Conference Paper

View full-text via PolyU eLinks SFX Query
Show full item record


Citations as of Feb 25, 2017

Page view(s)

Last Week
Last month
Checked on Aug 20, 2017

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.