A Multi-Hypothesis Approach for Off-Line Signature Verification with HMMs

In this paper, an approach based on the combination of discrete Hidden Markov Models (HMMs) in the ROC space is proposed to improve the performance of off-line signature verification (SV) systems designed from limited and unbalanced training data. This approach is inspired by the multiple-hypothesis principle, and allows the system to choose, 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 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's performance during training and significantly reduces the error rates during operations. The experiments performed by using a real-world SV database with random, simple and skilled forgeries, indicated that the proposed approach can reduce the average error rates by more than 17%.

[1]  Hiromichi Fujisawa.  Robustness Design of Industrial Strength Recognition Systems , 2007 .

[2]  Anil K. Jain,et al.  On-line signature verification, , 2002, Pattern Recognit..

[3]  A. El-Yacoubi,et al.  Off-line signature verification using HMMs and cross-validation , 2000, Neural Networks for Signal Processing X. Proceedings of the 2000 IEEE Signal Processing Society Workshop (Cat. No.00TH8501).

[4]  Arun Ross,et al.  Information fusion in fingerprint authentication , 2003 .

[5]  Mahesan Niranjan,et al.  Realisable Classifiers: Improving Operating Performance on Variable Cost Problems , 1998, BMVC.

[6]  Luiz Eduardo Soares de Oliveira,et al.  Off-line Signature Verification Using Writer-Independent Approach , 2007, 2007 International Joint Conference on Neural Networks.

[7]  Chris Drummond,et al.  Discriminative vs. Generative Classifiers for Cost Sensitive Learning , 2006, Canadian Conference on AI.

[8]  R. Austria.  Declaration , 1987 .

[9]  Flávio Bortolozzi,et al.  Off-line signature verification using HMM for random, simple and skilled forgeries , 2001, Proceedings of Sixth International Conference on Document Analysis and Recognition.

[10]  Tom Fawcett,et al.  ROC Graphs: Notes and Practical Considerations for Data Mining Researchers , 2003 .

[11]  Robert Sabourin,et al.  A Human-Centric Off-Line Signature Verification System , 2007, Ninth International Conference on Document Analysis and Recognition (ICDAR 2007).

[12]  Eric Granger,et al.  State of the Art in Off-Line Signature Verification , 2008 .

[13]  Francesco Tortorella,et al.  A ROC-based reject rule for dichotomizers , 2005, Pattern Recognit. Lett..