On the Use of Log-Likelihood Ratio Based Model-Specific Score Normalisation in Biometric Authentication
Conference paper
Abstract
It has been shown that the authentication performance of a biometric system is dependent on the models/templates specific to a user. As a result, some users may be more easily recognised or impersonated than others. We propose a model-specific (or user-specific) likelihood based score normalisation procedure that can reduce this dependency. While in its original form, such an approach is not feasible due to the paucity of data, especially of the genuine users, we stabilise the estimates of local model parameters with help of the user-independent (hence global) parameters. The proposed approach is shown to perform better than the existing known score normalisation procedures, e.g., the Z-, F- and EER-norms, in the majority of experiments carried out on the XM2VTS database . While these existing procedures are linear functions, the proposed likelihood based approach is quadratic but its complexity is further limited by a set of constraints balancing the contributions of the local and the global parameters, which are crucial to guarantee good generalisation performance.
Keywords
Gaussian Mixture Model Equal Error Rate Biometric System False Acceptance Rate False Rejection Rate
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
Download
to read the full conference paper text
References
- 1.Doddington, G., Liggett, W., Martin, A., Przybocki, M., Reynolds, D.: Sheep, Goats, Lambs and Woves: A Statistical Analysis of Speaker Performance in the NIST 1998 Speaker Recognition Evaluation. In: ICSLP. Int’l Conf. Spoken Language Processing, Sydney (1998)Google Scholar
- 2.Furui, S.: Cepstral Analysis for Automatic Speaker Verification. IEEE Trans. Acoustic, Speech and Audio Processing / IEEE Trans. on Signal Processing 29(2), 254–272 (1981)CrossRefGoogle Scholar
- 3.Pierrot, J.-B.: Elaboration et Validation d’Approaches en Vérification du Locuteur, Ph.D. thesis, ENST, Paris (September 1998)Google Scholar
- 4.Chen, K.: Towards Better Making a Decision in Speaker Verification. Pattern Recognition 36(2), 329–346 (2003)CrossRefGoogle Scholar
- 5.Saeta, J.R., Hernando, J.: On the Use of Score Pruning in Speaker Verification for Speaker Dependent Threshold Estimation. In: The Speaker and Language Recognition Workshop (Odyssey), Toledo, pp. 215–218 (2004)Google Scholar
- 6.Jonsson, K., Kittler, J., Li, Y.P., Matas, J.: Support vector machines for face authentication. Image and Vision Computing 20, 269–275 (2002)CrossRefGoogle Scholar
- 7.Lindberg, J., Koolwaaij, J.W., Hutter, H.-P., Genoud, D., Blomberg, M., Pierrot, J.-B., Bimbot, F.: Techniques for a priori Decision Threshold Estimation in Speaker Verification. In: Proc. of the Workshop Reconnaissance du Locuteur et ses Applications Commerciales et Criminalistiques (RLA2C), Avignon, pp. 89–92 (1998)Google Scholar
- 8.Genoud, D.: Reconnaissance et Transformation de Locuteur, Ph.D. thesis, Ecole Polythechnique Fédérale de Lausanne (EPFL), Switzerland (1998)Google Scholar
- 9.Fierrez-Aguilar, J., Ortega-Garcia, J., Gonzalez-Rodriguez, J.: Target Dependent Score Normalisation Techniques and Their Application to Signature Verification. In: Zhang, D., Jain, A.K. (eds.) ICBA 2004. LNCS, vol. 3072, pp. 498–504. Springer, Heidelberg (2004)Google Scholar
- 10.Auckenthaler, R., Carey, M., Lloyd-Thomas, H.: Score Normalization for Text-Independant Speaker Verification Systems. Digital Signal Processing (DSP) Journal 10, 42–54 (2000)CrossRefGoogle Scholar
- 11.Ben, M., Blouet, R., Bimbot, F.: A Monte-Carlo Method For Score Normalization in Automatic Speaker Verification Using Kullback-Leibler Distances. In: ICASSP. Proc. Int. Conf. Acoustics, Speech and Signal Processing, Orlando, vol. 1, pp. 689–692 (2002)Google Scholar
- 12.Poh, N., Bengio, S.: F-ratio Client-Dependent Normalisation on Biometric Authentication Tasks. In: ICASSP. IEEE Int’l Conf. Acoustics, Speech, and Signal Processing, Philadelphia, pp. 721–724 (2005)Google Scholar
- 13.Poh, N., Bengio, S.: Database, Protocol and Tools for Evaluating Score-Level Fusion Algorithms in Biometric Authentication. Pattern Recognition 39(2), 223–233 (2005)CrossRefGoogle Scholar
- 14.Toh, K.-A., Jiang, X., Yau, W.-Y.: Exploiting Global and Local Decision for Multimodal Biometrics Verification. IEEE Trans. on Signal Processing 52(10), 3059–3072 (2004)CrossRefGoogle Scholar
- 15.Gauvain, J.L., Lee, C.-H.: Maximum a Posteriori Estimation for Multivariate Gaussian Mixture Obervation of Markov Chains. IEEE Tran. Speech Audio Processing 2, 290–298 (1994)Google Scholar
- 16.Friedman, J.: Regularized discriminant analysis. J. American Statiscal Association 84, 165–175 (1989)CrossRefGoogle Scholar
- 17.Dass, S.C., Zhu, Y., Jain, A.K.: Validating a biometric authentication system: Sample size requirements. IEEE Trans. Pattern Analysis and Machine Intelligence 28(12), 1302–1319 (2006)CrossRefGoogle Scholar
- 18.Matas, J., Hamouz, M., Jonsson, K., Kittler, J., Li, Y., Kotropoulos, C., Tefas, A., Pitas, I., Tan, T., Yan, H., Smeraldi, F., Begun, J., Capdevielle, N., Gerstner, W., Ben-Yacoub, S., Abdeljaoued, Y., Mayoraz, E.: Comparison of Face Verification Results on the XM2VTS Database. In: Proc. 15th Int’l Conf. Pattern Recognition, Barcelona, vol. 4, pp. 858–863 (2000)Google Scholar
- 19.Martin, A., Doddington, G., Kamm, T., Ordowsk, M., Przybocki, M.: The DET Curve in Assessment of Detection Task Performance. In: Proc. Eurospeech 1997, Rhodes, pp. 1895–1898 (1997)Google Scholar
- 20.Bengio, S., Mariéthoz, J.: The Expected Performance Curve: a New Assessment Measure for Person Authentication. In: The Speaker and Language Recognition Workshop (Odyssey), Toledo, pp. 279–284 (2004)Google Scholar
Copyright information
© Springer-Verlag Berlin Heidelberg 2007