Abstract
The neural and statistical classifiers employed in off-line signature verification (SV) systems are often designed from limited and unbalanced training data. In this article, an approach based on the combination of discrete Hidden Markov Models (HMMs) in the ROC space is proposed to improve the performance of these systems. Inspired by the multiple-hypothesis principle, this approach allows the system to select, from a set of different HMMs, the most suitable solution for a given input sample. By training an ensemble of user-specific HMMs with different number of states and different codebook sizes, and then combining these models in the ROC space, it is possible to construct a composite ROC curve that provides a more accurate estimation of system performance. Moreover, in testing mode, the corresponding operating points—which may be selected dynamically according to the risk associated with input samples—can significantly reduce the error rates. Experiments performed by using a real-world off-line SV database, with random, simple and skilled forgeries, indicate that the multi-hypothesis approach can reduce the average error rates by more than 17%, as well as the number of HMM states by 48%.
Similar content being viewed by others
References
Batista L., Rivard D., Sabourin R., Granger E., Maupin P.: State of the art in off-line signature verification. In: Verma, B., Blumenstein, M. (eds) Pattern Recognition Technologies and Applications: Recent Advances, 1st edn, IGI Global, Hershey (2007)
Britto, A.: A two-stage HMM-based method for recognizing handwritten numeral strings. PhD thesis, PUC-PR, Brasil (2001)
Coetzer, H., Sabourin, R.: A human-centric off-line signature verification system. In: International Conference on Document Analysis and Recognition (ICDAR 2007), pp. 153–157 (2007)
Coetzer J., Herbst B., du Preez A.: Off-line signature verification using the discrete radon transform and a hidden markov model. EURASIP J. Appl. Signal Process. 4, 559–571 (2004)
Drummond, C.: Discriminative vs. generative classifiers for cost sensitive learning. In: Canadian Conference on Artificial Intelligence. Lecture Notes in Artificial Intelligence, pp. 479–490 (2006)
El-Yacoubi, A., Justino, E., Sabourin, R., Bortolozzi, F.: Off-line signature verification using hmms and cross-validation. In: IEEE Workshop on Neural Networks for Signal Processing, pp. 859–868 (2000)
Fawcett T.: An introduction to roc analysis. Mach. Learn. 27, 861–874 (2006)
Ferrer M., Alonso J., Travieso C.: Offline geometric parameters for automatic signature verification using fixed-point arithmetic. IEEE Trans. Pattern Anal. Mach. Intell. 27(6), 993–997 (2005)
Flach, P., Wu, S.: Repairing concavities in roc curves. In: Proceedings 2003 UK Workshop on Computational Intelligence, pp. 38–44, August 2003
Freitas, C., Morita, M., Oliveira, L., Justino, E., Yacoubi, A., Lethelier, E., Bortolozzi, F., Sabourin, R.: Bases de dados de cheques bancarios brasileiros. In: XXVI Conferencia Latinoamericana de Informatica (2000)
Fujisawa H.: Robustness Design of Industrial Strength Recognition Systems. Springer, New York (2007)
Gibbons J.: Nonparametric Statistical Inference, 2nd edn. M Dekker, New York (1985)
Gonzalez, R., Woods, R. (eds.): Digital Image Processing, 2nd edn. Prentice Hall, Upper Saddle River (2002)
Gotoh Y., Hochberg M., Silverman H.: Efficient training algorithms for HMMs using incremental estimation. IEEE Trans. Speech Audio Process. 6(6), 539–548 (1998)
Griess F., Jain A.: On-line signature verification. Pattern Recognit. 35, 2963–2972 (2002)
Huang, X., Acero, A., Hon, H. (eds.): Spoken Language Processing. Prentice Hall, Upper Saddle River (2001)
Impedovo D., Pirlo G.: Automatic signature verification: the state of the art. IEEE Trans. Syst. Man Cybern. C 38(5), 609–635 (2008)
Jain, A., Ross, A.: Learning user-specific parameters in a multibiometric system. In: International Conference on Image Processing (ICIP), pp. 57–60 (2002)
Justino, E.: O Grafismo e os Modelos Escondidos de Markov na Verificacao Automatica de Assinaturas. PhD thesis, PUC-PR, Brasil (2001)
Justino, E., El-Yacoubi, A., Bortolozzi, F., Sabourin, R.: An off-line signature verification system using hmm and graphometric features. In: International Workshop on Document Analysis Systems, pp. 211–222 (2000)
Justino, E., Bortolozzi, F., Sabourin, R.: Off-line signature verification using hmm for random, simple and skilled forgeries. In: International Conference on Document Analysis and Recognition, pp. 105–110 (2001)
Justino E., Bortolozzi F., Sabourin R.: A comparison of svm and hmm classifiers in the off-line signature verification. Pattern Recogn. Lett. 26(9), 1377–1385 (2005)
Makhoul J., Roucos S., Gish H.: Vector quantization in speech coding. IEEE 73, 1551–1558 (1985)
Murshed, N., Bortolozzi, F., Sabourin, R.: Off-line signature verification using fuzzy artmap neural networks. In: IEEE International Conference on Neural Networks, pp. 2179–2184 (1995)
Oliveira, L., Justino, E., Sabourin, R.: Off-line signature verification using writer-independent approach. In: International Joint Conference on Neural Networks (IJCNN), pp. 2539–2544, 2007
Plamondon R.: Progress in Automatic Signature Verification. Word Scientific, Singapore (1994)
Rabiner L.: A tutorial on hidden markov models and selected applications in speech recognition. IEEE 77(2), 257–286 (1989)
Rigoll, G., Kosmala, A.: A systematic comparison between on-line and off-line methods for signature verification with hidden markov models. In: International Conference on Pattern Recognition, vol. 2, pp. 1755–1757 (1998)
Scott M., Niranjan M., Prager R.: Realisable Classifiers: Improving Operating Performance on Variable Cost Problems. University of Southampton, UK (1998)
Tortorella F.: A ROC-based reject rule for dichotomizers. Pattern Recogn. Lett. 26, 167–180 (2005)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Batista, L., Granger, E. & Sabourin, R. Improving performance of HMM-based off-line signature verification systems through a multi-hypothesis approach. IJDAR 13, 33–47 (2010). https://doi.org/10.1007/s10032-009-0101-0
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10032-009-0101-0