Abstract
We present a calibration algorithm that converts biometric matching scores into probability-based confidence scores. Using the context of iris biometrics, we show – theoretically and by experiments – that in addition to attaching a meaningful confidence measure to the output, this calibration technique yields the best possible detection error trade-off (DET) curves, both at the score level and at the decision level, thus maximizing the overall performance of the biometric system.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Daugman, J.: How iris recognition works. IEEE Transactions on Circuits and Systems for Video Technology 14(1), 21–30 (2004)
Boström, H.: Maximizing the area under the ROC curve using incremental reduced error pruning. In: ICML 2005 Workshop on ROC Analysis in Machine Learning (2005)
Flach, P.A.: The geometry of ROC space: Understanding machine learning metrics through ROC isometrics. In: Intern. Conference on Machine Learning, pp. 194–201 (2003)
Gorodnichy, D.O.: Multi-order analysis framework for comprehensive biometric performance evaluation. In: SPIE Conference on Defense, Security, and Sensing (2010)
Gorodnichy, D.O., Hoshino, R.: Calibrated confidence scoring for biometric identification. In: NIST International Biometric Performance Conference, IBPC 2010 (2010)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2010 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Gorodnichy, D.O., Hoshino, R. (2010). Score Calibration for Optimal Biometric Identification. In: Farzindar, A., Kešelj, V. (eds) Advances in Artificial Intelligence. Canadian AI 2010. Lecture Notes in Computer Science(), vol 6085. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-13059-5_46
Download citation
DOI: https://doi.org/10.1007/978-3-642-13059-5_46
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-13058-8
Online ISBN: 978-3-642-13059-5
eBook Packages: Computer ScienceComputer Science (R0)