A performance improvement method for existing fingerprint systems

Authors:
Chunxiao Ren;Yilong Yin;Yanbin Ning
Affiliations:
Administration Center, Shandong Academy of Information & Communication Technology, Jinan, China;School of Computer Science and Technology, Shandong University, Jinan, P. R. China;School of Computer Science and Technology, Shandong University, Jinan, P. R. China
Venue:
CCBR'12 Proceedings of the 7th Chinese conference on Biometric Recognition
Year:
2012

Citing 5
Cited 0

FVC2000: Fingerprint Verification Competition

IEEE Transactions on Pattern Analysis and Machine Intelligence
FVC2002: Second Fingerprint Verification Competition

ICPR '02 Proceedings of the 16 th International Conference on Pattern Recognition (ICPR'02) Volume 3 - Volume 3
Handbook of Multibiometrics (International Series on Biometrics)

Handbook of Multibiometrics (International Series on Biometrics)
Pattern Recognition and Machine Learning (Information Science and Statistics)

Pattern Recognition and Machine Learning (Information Science and Statistics)
A comparative study of combining multiple enrolled samples for fingerprint verification

Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

How to improve the performance of an existing fingerprint system is an interesting and meaningful problem. Considering the widespread deployment of fingerprint systems, the performance improvement method is very practical and instructive to not only the existing fingerprint systems but also the next biometric systems. In this paper, we propose a novel performance improvement method based on fingerprint's Möbius representation and Choquet integral for an existing fingerprint system. The basic idea of our method is to map the fingerprint similarity as a distance in a geometric space firstly, and then transform the similarity problem between the impressions into a geometric problem through using multiple impressions, and last, map the results obtained in geometric space back to solve the fingerprint similarity problem. The experiments show that the performance achieved by using this method is better than that of other methods.