A novel modality independent score-level quality measure

The performance of biometric systems deployed in real world scenarios is affected adversely by the degradation in the input quality. This degradation can take place either in the input or due to inherent design of the deployed system. Quality Measures aim to provide a quantitative indication of this degradation which can be utilized to help access the probability that the provided biometric decision is correct. In this paper we present a novel, modality independent, score level quality measure based on Mahalanobis Distance. This measure exploits the inter-dependencies which exist between the data and matching score to reduce the error rates and improve the performance of the deployed biometric system. The effectiveness of the proposed approach is verified by extensive testing using the standard databases available of fingerprints and Iris data.

[1]  Samy Bengio,et al.  Improving Fusion with Margin-Derived Confidence in Biometric Authentication Tasks , 2005, AVBPA.

[3]  John Daugman,et al.  High Confidence Visual Recognition of Persons by a Test of Statistical Independence , 1993, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  Julian Fiérrez,et al.  Biosec baseline corpus: A multimodal biometric database , 2007, Pattern Recognit..

[5]  Libor Masek,et al.  MATLAB Source Code for a Biometric Identification System Based on Iris Patterns , 2003 .

[6]  Venu Govindaraju,et al.  K-plet and Coupled BFS: A Graph Based Fingerprint Representation and Matching Algorithm , 2006, ICB.

[7]  Lawrence M Wein,et al.  Using fingerprint image quality to improve the identification performance of the U.S. Visitor and Immigrant Status Indicator Technology Program. , 2005, Proceedings of the National Academy of Sciences of the United States of America.

[8]  Mark C. Huggins,et al.  Confidence metrics for speaker identification , 2002, INTERSPEECH.

[9]  P. Mahalanobis.  On the generalized distance in statistics , 1936 .

[10]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[11]  Josef Kittler,et al.  On Combination of Face Authentication Experts by a Mixture of Quality Dependent Fusion Classifiers , 2007, MCS.

[12]  Elham Tabassi,et al.  Performance of Biometric Quality Measures , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  Julian Fiérrez,et al.  Multimodal biometric authentication using quality signals in mobile communications , 2003, 12th International Conference on Image Analysis and Processing, 2003.Proceedings..