ON THE EFFECT OF SCORE EQUALIZATION IN SVM MULTIMODAL BIOMETRIC SYSTEMS

Pascual Ejarque, Javier Hernando

2007

Abstract

Most Support Vector Machine (SVM) based systems make use of conventional methods for the normalization of the features or the scores previously to the fusion stage. In this work, in addition to the conventional methods, two equalization methods, histogram equalization, which was recently introduced in multimodal systems, and Bi-Gaussian equalization, which is presented in this paper, are applied upon the scores in a multimodal person verification system composed by prosodic, speech spectrum, and face information. The equalization techniques have obtained the best results; concretely, Bi-Gaussian equalization outperforms in more than a 22.19 % the results obtained by Min-Max normalization, the most used normalization technique in SVM fusion systems. The prosodic and speech spectrum scores have been provided by speech experts using records of the Switchboard I database and the face scores have been obtained by a face recognition system upon XM2VTS database.

References

  1. Bolle, R. M., Connell, J. H., Pankanti, S., Ratha, N. K., and Senior, A. W., 2004. Guide to Biometrics, Springer-Verlag New York, Inc.
  2. Jain, A. K., Nandakumar, K., and Ross, A., 2005. Score normalization in multimodal biometric systems. Pattern Recognition, vol. 38, no. 12, pp. 2270-2285.
  3. Cristianini, N., and Shawe-Taylor, J., 2000. An introduction to support vector machines (and other kernel-based learning methods), Cambridge University Press.
  4. Wan, V., and Renals, S., 2005. Speaker verification using sequence discriminant support vector machines. IEEE Trans. on Speech and Audio Processing, 13:203-210.
  5. Farrús, M., Garde, A., Ejarque, P., Luque, J., and Hernando, J., 2006. On the Fusion of Prosody, Voice Spectrum and Face Features for Multimodal Person Verification. Proc. of Interspeech 2006, Pittsburgh, USA.
  6. Ejarque, P., Garde, A., Anguita, J., and Hernando, J., 2007. On the use of genuine-impostor statistical information for score fusion in multimodal biometrics. Annals of Telecommunication, Multimodal Biometrics, vol 62, nº 1-2.
  7. Wolf, J. J., 1972. Efficient acoustic parameters for speaker recognition. Journal of the Acoustical Society of America, vol. 51, pp. 2044-2056.
  8. Nadeu, C., Mariño, J. B., Hernando, J., and Nogueiras, A., 1996. Frequency and time-filtering of filter-bank energies for HMM speech recognition. ICSLP, pp. 430-433, Philadelphia, USA.
  9. Tefas, A., Zafeiriou, S., and Pitas, I., 2005. Discriminant NMFfaces for frontal face verification. Proc. of IEEE International Workshop on Machine Learning for Signal Processing (MLSP 2005), Mystic, Connecticut, September 28-30.
  10. Jain, A., 1986. Fundamentals of Digital Image Processing, Prentice-Hall, pp 241 - 243.
  11. Balchandran, R., and Mammone, R., 1998. Non parametric estimation and correction of non-linear distortion in speech systems. Proc. IEEE Int. Conf. Acoust. Speech Signal Proc.
  12. Godfrey, J. J., Holliman, E. C., and McDaniel, J., 1990. Switchboard: Telephone speech corpus for research and development. ICASSP.
  13. Lüttin, J., and Maître, G., 1998. Evaluation Protocol for the Extended M2VTS Database (XM2VTSDB). IDIAP Communication 98-05, Martigny, Switzerland.
Download


Paper Citation


in Harvard Style

Ejarque P. and Hernando J. (2007). ON THE EFFECT OF SCORE EQUALIZATION IN SVM MULTIMODAL BIOMETRIC SYSTEMS . In Proceedings of the Second International Conference on Security and Cryptography - Volume 1: SECRYPT, (ICETE 2007) ISBN 978-989-8111-12-8, pages 33-38. DOI: 10.5220/0002130000330038


in Bibtex Style

@conference{secrypt07,
author={Pascual Ejarque and Javier Hernando},
title={ON THE EFFECT OF SCORE EQUALIZATION IN SVM MULTIMODAL BIOMETRIC SYSTEMS},
booktitle={Proceedings of the Second International Conference on Security and Cryptography - Volume 1: SECRYPT, (ICETE 2007)},
year={2007},
pages={33-38},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002130000330038},
isbn={978-989-8111-12-8},
}


in EndNote Style

TY - CONF
JO - Proceedings of the Second International Conference on Security and Cryptography - Volume 1: SECRYPT, (ICETE 2007)
TI - ON THE EFFECT OF SCORE EQUALIZATION IN SVM MULTIMODAL BIOMETRIC SYSTEMS
SN - 978-989-8111-12-8
AU - Ejarque P.
AU - Hernando J.
PY - 2007
SP - 33
EP - 38
DO - 10.5220/0002130000330038