Overview of the Face Recognition Grand Challenge
CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 1 - Volume 01
Performance Evaluation of Fingerprint Verification Systems
IEEE Transactions on Pattern Analysis and Machine Intelligence
Empirical Studies of the Existence of the Biometric Menagerie in the FRGC 2.0 Color Image Corpus
CVPRW '06 Proceedings of the 2006 Conference on Computer Vision and Pattern Recognition Workshop
IEEE Transactions on Systems, Man, and Cybernetics, Part A: Systems and Humans - Special issue on recent advances in biometrics
Electronic Commerce Research and Applications
| Hi-index | 0.00 |
The intersubject dependencies of false nonmatch rates were investigated for a minutiae-based biometric authentication process using single enrollment and verification measurements. A large number of genuine comparison scores were subjected to statistical inference tests that indicated that the number of false nonmatches depends on the subject and finger under test. This result was also observed if subjects associated with failures to enroll were excluded from the test set. The majority of the population (about 90%) showed a false nonmatch rate that was considerably smaller than the average false nonmatch rate of the complete population. The remaining 10% could be characterized as "goats" due to their relatively high probability for a false nonmatch. The image quality reported by the template extraction module only weakly correlated with the genuine comparison scores. When multiple verification attempts were investigated, only a limited benefit was observed for "goats," since the conditional probability for a false nonmatch given earlier nonsuccessful attempts increased with the number of attempts. These observations suggest that (1) there is a need for improved identification of "goats" during enrollment (e.g., using dedicated signal-driven analysis and classification methods and/or the use of multiple enrollment images) and (2) there should be alternative means for identity verification in the biometric system under test in case of two subsequent false nonmatches.