Systems and Means of Informatics

November 2013, Volume 23, Issue 2, pp 22-34

TOPOLOGICAL MODEL OF FINGERPRINT IMAGE

  • V. Yu. Gudkov
  • O. S. Ushmaev

Abstract

The paper describes new fingerprint image model that consist of topological and metric features of minutiae neighborhoods. In order to create model, we propose to calculate topological features in neighborhood of each fingerprint minutiae: ridge ending or bifurcation. Topological feature is interrelation between two minutiae. Simple example of topological feature is two minutiae are situated on one ridge. Then list of topological features is constructed. Description of each feature is extended with metric information: distance between minutiae. The proposed model further is used for fingerprint identification. Experiments show that simultaneous usage of topological and metric features significantly improves accuracy of fingerprint identification. On public available FVC2004 dataset FRR@FAR=10 is 2%.

[+] References (11)

[+] About this article