A Novel Method for Multibiometric Fusion Based on FAR and FRR

Authors:
Yong Li;Jianping Yin;Jun Long;En Zhu
Affiliations:
School of Computer, National University of Defense Technology, Changsha, China 410073;School of Computer, National University of Defense Technology, Changsha, China 410073;School of Computer, National University of Defense Technology, Changsha, China 410073;School of Computer, National University of Defense Technology, Changsha, China 410073
Venue:
MDAI '09 Proceedings of the 6th International Conference on Modeling Decisions for Artificial Intelligence
Year:
2009

Citing 6
Cited 0

On Combining Classifiers

IEEE Transactions on Pattern Analysis and Machine Intelligence
Comparison of Face Verification Results on the XM2VTS Database

ICPR '00 Proceedings of the International Conference on Pattern Recognition - Volume 4
Handbook of Multibiometrics (International Series on Biometrics)

Handbook of Multibiometrics (International Series on Biometrics)
Database, protocols and tools for evaluating score-level fusion algorithms in biometric authentication

Pattern Recognition
Likelihood Ratio-Based Biometric Score Fusion

IEEE Transactions on Pattern Analysis and Machine Intelligence
Biometric recognition: overview and recent advances

CIARP'07 Proceedings of the Congress on pattern recognition 12th Iberoamerican conference on Progress in pattern recognition, image analysis and applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

Based on the fusion of multiple biometric sources, Multibiometric systems can be expected to be more accurate due to the presence of multiple pieces of evidence. Multibiometric system design is a challenging problem because it is very difficult to choose the optimal fusion strategy. Score level fusion is the most commonly used approach in Multibiometric systems. The distribution of genuine and imposter scores are very important for score fusion of Multibiometric systems. FRR (False Reject Rate) and FAR (False Accept Rate) are two key parameters to cultivate the distribution of genuine and imposter scores. In this paper, we first present a model for Multibiometric fusion and then proposed a novel approach for score level fusion which is based on FAR and FRR. By this method, the match scores first are transformed into LL1s and then the sum rule is used to combine the LL1s of the scores. The experimental results show that the new fusion scheme is efficient for different Multibiometric systems.