Munich, Mario E. and Perona, Pietro (2003) Visual identification by signature tracking. IEEE Transactions on Pattern Recognition and Machine Intelligence, 25 (2). pp. 200-217. ISSN 0162-8828. http://resolver.caltech.edu/CaltechAUTHORS:MUNieeetpami03
- Published Version
See Usage Policy.
Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:MUNieeetpami03
We propose a new camera-based biometric: visual signature identification. We discuss the importance of the parameterization of the signatures in order to achieve good classification results, independently of variations in the position of the camera with respect to the writing surface. We show that affine arc-length parameterization performs better than conventional time and Euclidean arc-length ones. We find that the system verification performance is better than 4 percent error on skilled forgeries and 1 percent error on random forgeries, and that its recognition performance is better than 1 percent error rate, comparable to the best camera-based biometrics.
|Additional Information:||© 2003 IEEE. Reprinted with permission. Manuscript received 24 Aug. 1 2001; revised 17 May 2002; accepted 14 July 2002. Posted online: 2003-02-19. Recommended for acceptance by S. Sclaroff. The authors gratefully acknowledge support from the US National Science Foundation Engineering Research Center on Neuromorphic Systems Engineering at Caltech (NSF) Cooperative Agreement No. EEC-9402726). They would also like to express their gratitude to all the subjects that collaborated in the experiments by providing their time and their signatures to build the example databases.|
|Subject Keywords:||Systems and applications, active and real-time vision, signature verification, signature recognition, biometrics|
|Usage Policy:||No commercial reproduction, distribution, display or performance rights in this work are provided.|
|Deposited By:||Archive Administrator|
|Deposited On:||14 Mar 2006|
|Last Modified:||28 Mar 2017 16:10|
Repository Staff Only: item control page