FeatureFingerprint verification competition 2006
Section snippets
Databases
Four databases created using three different scanners and the SFinGe synthetic generator were used in the FVC2006 benchmark (see Table 2). Figure 1 shows an example image at the same scale factor from each database.
Different from the first three editions of FVC, data collection in FVC2006 was performed without deliberately introducing difficulties such as exaggerated distortion, large amounts of rotation and displacement, wet/dry impressions, etc, but the population is more heterogeneous and
Testing protocol
Participants submitted each algorithm in the form of two executable programs: the first for enrolling a fingerprint image and producing the corresponding template, and the second for comparing a fingerprint template to a fingerprint image and producing a comparison score in the range [0,1]. The input includes a database-specific configuration file. Each algorithm is tested by performing, for each database, the following comparisons:
- •
Genuine recognition attempts: the template of each
Results
The following performance indicators were measured and made available at the FVC2006 web site (http://bias.csr.unibo.it/fvc2006/)
- •
Genuine and impostor score histograms;
- •
False Match Rate (FMR) and False Non-Match Rate (FNMR) graphs and Decision Error Trade-off (DET) graph;
- •
Failure-to-Enrol Rate (FTE) and Failure-to-Compare Rate (FTC) which can be reported by the algorithm or imposed by the test procedure in case of timeout, program crash, exceeded memory limit, exceeded template limit or
Conclusions
Performance evaluation is important for all pattern recognition applications and particularly so for biometrics, which is receiving widespread international attention for citizen identity verification and identification in large-scale applications. Unambiguously and reliably assessing the current state of the technology is essential for understanding its limitations and addressing future research requirements.
The interest shown in the FVC testing program by algorithm developers continues to be
References (0)
Cited by (156)
A secure fingerprint template generation mechanism using visual secret sharing with inverse halftoning
2023, Journal of Visual Communication and Image RepresentationZJUT-EIFD: A Synchronously Collected External and Internal Fingerprint Database
2024, IEEE Transactions on Pattern Analysis and Machine IntelligenceAdvancing Fingerprint Template Generation and Matching With Recast Minutiae Clustering and mRBFN
2024, Advances in Artificial Intelligence and Machine LearningHigh Precision Fingerprint Verification for Small Area Sensor Based on Deep Learning
2024, IEICE Transactions on Fundamentals of Electronics, Communications and Computer SciencesConsiderations on the Evaluation of Biometric Quality Assessment Algorithms
2024, IEEE Transactions on Biometrics, Behavior, and Identity Science
This feature was provided by Raffaele Cappelli, Matteo Ferrara, Annalisa Franco and Davide Maltoni of the Biometric System Laboratory, University of Bologna.