Multilevel fusion for fast online signature recognition using multi-section VQ and time modelling

Signature recognition is one of the most important biometrics authentication methods, is an integral part of current business activities, and is considered a non-invasive and non-threatening process. This paper presents an online signature verification system using multi-section VQ. We have used multi-section codebooks for signature recognition by splitting the signature into several sections with every section having its own codebook. The final result is based on the score level fusion of the results of each codebook. Moreover, multilevel fusion is performed in this trial to improve the accuracy. We have used SVC database that contains skilled forgery samples. Our experimental results on SVC database have shown 100 % accuracy with 0.003 EER.

[1]  Loris Nanni,et al.  Ensemble of Parzen window classifiers for on-line signature verification , 2005, Neurocomputing.

[2]  Emile H. L. Aarts,et al.  On-line signature verification with hidden Markov models , 1998, Proceedings. Fourteenth International Conference on Pattern Recognition (Cat. No.98EX170).

[3]  Carlos Vivaracho-Pascual,et al.  An efficient low cost approach for on-line signature recognition based on length normalization and fractional distances , 2009, Pattern Recognit..

[4]  Marcos Faúndez-Zanuy,et al.  Fast on-line signature recognition based on VQ with time modeling , 2011, Eng. Appl. Artif. Intell..

[5]  Bernadette Dorizzi,et al.  On Using the Viterbi Path Along With HMM Likelihood Information for Online Signature Verification , 2007, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[6]  Bernhard Sick,et al.  Online Signature Verification With Support Vector Machines Based on LCSS Kernel Functions , 2010, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[7]  Nalini K. Ratha,et al.  Proceedings of the 5th international conference on Audio- and Video-Based Biometric Person Authentication , 2005 .

[8]  Marcos Faúndez-Zanuy,et al.  On-line signature recognition based on VQ-DTW , 2007, Pattern Recognit..

[9]  Michael C. Fairhurst,et al.  Biosecure reference systems for on-line signature verification: A study of complementarity , 2007, Ann. des Télécommunications.

[10]  Marcos Faúndez-Zanuy,et al.  Efficient on-line signature recognition based on multi-section vector quantization , 2010, Pattern Analysis and Applications.

[11]  Gerhard Rigoll,et al.  A systematic comparison between on-line and off-line methods for signature verification with hidden Markov models , 1998, Proceedings. Fourteenth International Conference on Pattern Recognition (Cat. No.98EX170).

[12]  Loris Nanni,et al.  An On-Line Signature Verification System Based on Fusion of Local and Global Information , 2005, AVBPA.

[13]  Hong Chang,et al.  SVC2004: First International Signature Verification Competition , 2004, ICBA.

[14]  Loris Nanni,et al.  Experimental comparison of one-class classifiers for online signature verification , 2006, Neurocomputing.

[15]  Hussein Zedan,et al.  Rough set approach to online signature identification , 2011, Digit. Signal Process..

[16]  Jonas Richiardi,et al.  Local and global feature selection for on-line signature verification , 2005, Eighth International Conference on Document Analysis and Recognition (ICDAR'05).

[17]  Michael R. Lyu,et al.  Spectrum Analysis Based onWindows with Variable Widths for Online Signature Verification , 2006, 18th International Conference on Pattern Recognition (ICPR'06).