Fingerprint Authentication Based on Matching Scores with Other Data

A method of person authentication based on matching scores with the fingerprint data of others is proposed. Fingerprint data of others is prepared in advance as a set of representative data. Input fingerprint data is verified against the representative data, and the person belonging to the fingerprint is confirmed from the set of matching scores. The set of scores can be thought of as a feature vector, and is compared with the feature vector already enrolled. In this paper, the mechanism of the proposed method, the person authentication system using this method are described, and its advantage. Moreover, the simple criterion and selection method of the representative data are discussed. The basic performance when general techniques are used for the classifier is FNMR-3.6% at FMR-0.1%.

[1]  Yoav Freund,et al.  Experiments with a New Boosting Algorithm , 1996, ICML.

[2]  Koichi Sasakawa,et al.  Personal verification system with high tolerance of poor-quality fingerprints , 1991, Other Conferences.

[3]  Bernard Zenko,et al.  Is Combining Classifiers Better than Selecting the Best One , 2002, ICML.

[4]  Bernard Zenko,et al.  Is Combining Classifiers with Stacking Better than Selecting the Best One? , 2004, Machine Learning.

[5]  João Gama,et al.  Cascade Generalization , 2000, Machine Learning.

[6]  Leo Breiman,et al.  Bagging Predictors , 1996, Machine Learning.

[7]  David H. Wolpert,et al.  Stacked generalization , 1992, Neural Networks.

[8]  Nalini K. Ratha,et al.  Enhancing security and privacy in biometrics-based authentication systems , 2001, IBM Syst. J..

[9]  Andy Adler.  Sample images can be independently restored from face recognition templates , 2003, CCECE 2003 - Canadian Conference on Electrical and Computer Engineering. Toward a Caring and Humane Technology (Cat. No.03CH37436).