An efficient scheme for dynamic signature verification

Personal signatures are easily forged because their verification is usually limited to a visual image comparison. This paper presents a dynamic signature verification system. The system analyzes signatures dynamically by considering their shape, time domain characteristics, such as speed and acceleration, and force domain characteristics, i.e. applied pressure. Then it compares these parameters with those of previously obtained master signatures. The results are converted into a percentage match figure to determine whether the signature is qualified as authentic or forgery. Experimental results show 92% authentic signature detection accuracy and 100% forgery signature detection accuracy. This high level of accuracy plus low computation requirements for analysis, make this system a commercially viable solution to the signature identification problem.

[1]  Azriel Rosenfeld.  Image analysis and computer vision: 1993 , 1994 .

[2]  Réjean Plamondon,et al.  Segmenting Handwritten Signatures at Their Perceptually Important Points , 1993, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  Azriel Rosenfeld,et al.  Image Analysis and Computer Vision: 1995 , 1996, Comput. Vis. Image Underst..

[4]  M. Khalil,et al.  Issues pertaining to optimal performance of an on-line autoregressive model based signature verification system , 1993, Proceedings of 36th Midwest Symposium on Circuits and Systems.

[5]  Ching Y. Suen,et al.  The State of the Art in Online Handwriting Recognition , 1990, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  Y. Sato,et al.  Online Signature Verification Based on Shape, Motion, and Writing Pressure , 1982 .

[7]  C. N. Liu,et al.  Automatic Signature Verification: System Description andField TestResults , 1979 .

[8]  Azriel Rosenfeld,et al.  Instrument grasp: a model and its effects on handwritten strokes , 1994, Pattern Recognit..

[9]  M. J. Paulik,et al.  A 1-D, sequence decomposition based, autoregressive hidden Markov model for dynamic signature identification and verification , 1993, Proceedings of 36th Midwest Symposium on Circuits and Systems.

[10]  Marc Parizeau,et al.  A Comparative Analysis of Regional Correlation, Dynamic Time Warping, and Skeletal Tree Matching for Signature Verification , 1990, IEEE Trans. Pattern Anal. Mach. Intell..

[11]  Azriel Rosenfeld,et al.  Linear feature compatibility for pattern-matching relaxation , 1992, Pattern Recognit..

[12]  Réjean Plamondon,et al.  A complexity measure of handwritten curves: modeling of dynamic signature forgery , 1993, IEEE Trans. Syst. Man Cybern..