Studying Of Multimodal Biometric System Using Different Svm Kernels
Keywords:
Binary classifiers, support vector machine, biometric system, feature extraction, fusion classifierAbstract
Biometric security system is based on a binary classification problem: reject or accept identity claim. The performance and robustness of biometric security systems can be improved by combining two or more different modalities (speech, face, fingerprint, etc.). In literature, one of the issues is to improve the performance of multimodal biometric system. In this paper, we investigated the performance of multimodal biometric system using different SVM kernel functions and its parameters. Our experiment on XM2VTS dataset shows systems have been developed based on different biometrics characteristics (fingerprint, face, speech, iris etc.) which help to distinguish individuals from each other. Each biometric has its own advantages and drawbacks due to its discriminative power, complexity, robustness involved. Research in last few years has shown that no single modal biometric system can achieve 100% authentication accuracy. This problem can be alleviated by combining two or more biometric modalities [1], also known as the field of multimodal biometric authentication. A user verification scenario involving two modalities (Face and Speech) [2] is figured out in Fig. 1. that the performance of multimodal biometric system is improved with the proper choice of kernel function and its parameters.
References
J. Kittler, “Combining classifiers: Atheoretical framework,” Pattern Anal. Applicat., vol. 1, pp. 18–27, 1998.
Ben-Yacoub, S.; Abdeljaoued, Y.; Mayoraz, “Fusion of Face and Speech Data for Person Identity Verification”,
Neural Networks, IEEE Transactions on Volume 10, Issue 5, Sep 1999 Page(s):1065 – 1074.
C. Bishop. Neural Networks for Pattern Recognition. Oxford University Press, 1999.
R. Brunelli and D. Falavigna, “Person identification using multiple cues,” IEEE Trans. Pattern Anal. Machine Intell., vol. 17, pp. 955–966, Oct. 1995.
S. Ikbal, H. Misra, and H. Bourlard. Phase Auto Correlation (PAC) derived Robust Speech Features. In Proc. IEEE Int'l Conf. Acoustics, Speech, and Signal Processing (ICASSP-03), pages 133{136, Hong Kong, 2003.
S. Marcel and S. Bengio. Improving Face Veri_cation Using Skin Color Information. In Proc. 16th Int. Conf. on Pattern Recognition, page unknown, Quebec, 2002.
K. K. Paliwal. Spectral Subband Centroids Features for Speech Recognition. In Proc. Int. Conf. Acoustics,
V.N. Vapnik, The Nature of Statistical Learning Theory, Springer, Berlin Heidelberg, New York, 1995.
N. Poh, C. Sanderson, and S. Bengio. An Investigation of Spectral Subband Centroids For Speaker Authentication. In Springer LNCS-3072, Int'l Conf. on Biometric Authentication (ICBA), pages 631-639, Hong Kong, 2004.
C. Corts and V.N. Vapnik, “Support Vector Networks,” Machine Learning , Vol. 20, pp. 273-297, 1995.
B. Scholkopf and A. J. Smola, Learning with kernels. MIT Press, Cambridge, MA, 2002.
L. Rabiner and B-H Juang. Fundamentals of Speech Recognition. Oxford University Press, 1993.
J. R. Quinlan, C4.5: Programs for Machine Learning. San Mateo, CA: Morgan Kaufmann, 1993.
P. Werbos, “Beyond regression: New tools for prediction and analysis in the behavioral sciences,” Ph.D. dissertation, Harvard Univ., Cambridge, MA, 1974.
D. E. Rumelhart, G. E. Hinton, and R. J. Williams, “Learning representations by back-propagating errors,” Nature, vol. 323, pp. 533–536, 1986.
K. Messer, J. Matas, J. Kittler, J. L¨uttin, and G. Maˆýtre, “XM2VTSDB: The extended M2VTS database,” in Proc. 2nd Int. Conf. Audio-Video Based Biometric Person Authentication, Washington, D.C., Mar. 22–23, 1999, pp. 72–77.
Fawcett, T. (2001). Using rule sets to maximize ROC performance. In Proceedings of the IEEE International Conference on Data Mining (ICDM-2001), pp. 131-138.
Norman Poh, Samy Bengio, “Database, Protocol and tools for evaluating score-level fusion algorithms in biometric authentication” IDIAP-RR 04-44, Switzerland, 2004.
J Kittler and A. Hojjatoleslami, “ Aweighted combination of c las s ifier s employ ing shared and di s tinc t representations”, in IEEE Proc. Comput. Vision Pattern Recognition, pp 924-929, 1998.
T. Joachims, “Making large Scale SVM learning Practical. Advances in Kernel Methods – Support Vector Learning, B. Scholkopf and C. Burges and A. Smola (ed.), MIT Press, 1999.
R. Fletcher, Practical Methods of Optimization. John Wiley & Sons, Inc., 2nd edition, 1987.
Shigeo Abe, Support Vector Machines for Pattern Classification, Springer, Kobe, Japan.
M. M. Fadel, R. K. Agrawal, Multimodal Biometric Authentication System: A Comparative Study of Fusion Classifier, Accepted for publishing in Optima-2007 National Conference, New Delhi, 2007.
B.Gutschoveen and P. Verlinde“Multi-model Identity Verification using Support Vector Machines (SVM)”, ISIF, 2000.
Downloads
Published
Issue
Section
License
Copyright (c) 2022 Trinity Journal of Management, IT & Media (TJMITM)
This work is licensed under a Creative Commons Attribution 4.0 International License.
