Document Type : Full Research Paper


1 Instructor, Department of Bioelectric, Biomedical Engineering School, Science and Research Branch, Islamic Azad University

2 Assistant Professor, Department of Bioelectric, Biomedical Engineering School, Amir Kabir University

3 Associate Professor, Department of Bioelectric, Biomedical Engineering School, Amir Kabir University



Dynamic signature verification based on temporal features are more precise than the static methods because in addition to position information of the drawing pattern, it uses local and global features extracted from velocity, acceleration, pressure and pen angle signals, while static methods only use image information. In this study, we segmented the signature patterns using the basic role of velocity in the control process of skilled movements and then the function features were extracted. In order to signal the matching evaluation, we applied five generalized functions and five weighting strategies for score level fusion. The results showed that the correlation criterion had the minimum error. The experiments on the database, consisting of persons of Persian, Chinese and English, showed that the skilled forgeries obtained an equal error rate (EER) of 0.87% and 1.24% for the user and universal thresholds, respectively. 


Main Subjects

[1]     Plamondon R., Lorette G., Automatic Signature Verification and Writer Identification - The State of the Art; Pattern Recognition; 1989; 22(2): 107-131.
[2]     Nalwa V.S., Automatic On-line Signature Verification; Proceedings of the IEEE; 1997; 85(2): 213-239.
[3]     Parizeau M., Plamondon R., A Comparative Analysis of Regional Correlation, Dynamic Time Warping, and Skeletal tree matching for Signature Verification; IEEE Trans. on Pattern Analysis and Machine Intelligence; 1990; 12(7): 710-717.
[4]     Lee L.L., Neural approaches for human signature verification; In Third International Conference on Signal Processing Proceedings; 1996: 1346-1349.
[5]     Yang L., Widjaja B.K., Prasad R., Application of Hidden Markov Models for Signature Verification; Pattern Recognition; 1995;28(2): 161-170.
[6]     Lejtman D.Z., George S.E., On-line handwritte signature verification using wavelets and backpropagation neural networks; Proceedings on the Sixth International Conference on Document Analysis and Recognition; 2001: 992-996.
[7]     Plamondon R., Alimi A.M., Yergeau P., Leclerc F., Modeling velocity profiles of rapid movements: a comparative study; Bilogical Cybernetics; 1993; 69: 119-128.
[8]     Stein R.B, What muscle variable(s) does the nervous system control in limb movements?; The Behavioral and Brain Science; 1982; 5: 535-577.
[9]     Plamondon R., A kinematic theory of rapid human movements: 1 Movement representation and generation; Biological Cybernetics; 1995(a); 72: 295- 307.
[10] Guerfali W., Plamondon R., The delta-lognormal theory for the generation and modeling of cursive characters; Proceedings of the International Conference on Document Analysis and Recognition; 1995:495-498.
[11] SVC; The First International Signature Verification Competition;
[12] رشیدی سعید؛ براورد کمی فرایند یادگیری مطلوب الگوهای حرکتی دست به کمک مدلسازی در افراد سالم و دارای برخی ضایعات عصبی-عضلانی، پایان‌نامه کارشناسی ارشد مهندسی پزشکی، دانشگاه صنعتی امیرکبیر، 1377.
[13] Salvador S., Chan M., Fast DTW: Toward Accurate Dynamic Time Warping in Linear Time and Space; Proc. On Advanced Conf. on Knowledge Discovery and Data Mining, Seattle, USA; 2004: 70-80.
[14] Chu S., Keogh E., Hart D., Pazzani M., Iterative Deepening Dynamic Time Warping for Time Series; In Proc. of the Second SIAM Intl. Conf. on Data Mining; Arlington; Virginia; 2002: 237-246.
[15] Feng H., Wah C., Online Signature Verification Using a New Extreme Points Warping Technique; Pattern Recognition Letters; 2003; 24(16): 2943-2951.
[16] Yeung D., Chang H., Xiong Y., George S., Kashi R., Matsumoto T., Rigoll G., SVC2004 First International Signature Verification Competotion; Lecture Notes in Computer Science, Springer-Verlag; 2004; 3072: 16- 22.
[17] Fierrez-Aguilar J., Adapted Fusion Schemes for Multimodal Biometric Authentication; PHD Thesis , Univ. Madrid, 2006.
[18] Doroz R., Porwik P., Para T., Wrobel K., Dynamic Signature Recognition Based on Velocity Change of Some Features; Int. J. Biometrics; 2008; 1(1): 47-62.