Determination of Decision Boundaries for Online Signature Verification
We are developing methods for online (or dynamical) signature verification. Our method is first to move the test signature to the sample signature so that the DP matching can be done well, and then compare the pen information along the matched points of the signatures. However, it is not easy to determine how to use the several elements of discrepancy between them. In this paper, we propose a verification method based on the discrimination by RBF network.
KeywordsDecision Boundary Template Data Online Signature Probability Density Function Model Template Signature
Unable to display preview. Download preview PDF.
- 2.Minsky, M.L., Papert, S.A.: Perceptrons – Expanded Edition. MIT, Cambridge (1969)Google Scholar
- 5.Tanaka, M., Ishino, Y., Shimada, H., Inoue, T.: DP Matching Using Kalman Filter as Pre-processing in On-line Signature Verification. In: 8th International Workshop on Frontiers in Handwriting Recognition, Niagara-on the-Lake, Ontario, Canada, August 6-8, pp. 502–507 (2002)Google Scholar
- 6.Tanaka, M., Ishino, Y., Shimada, H., Inoue, T.: Dynamical Scaling in Online Hand-written Characters’ Matching. In: 9th International Conference on Neural Information Processing, Singapore, November 19-22, 5 pages (CD ROM) (2002)Google Scholar
- 7.Tanaka, M., Ishino, Y., Shimada, H., Inoue, T., Bargiela, A.: Analysis of Iterative Procedure of Matching Two drawings by DP Matching and Estimation of Time-Variant Transformation Parameters. In: The 34th International Symposium on Stochastic Systems Theory and Its Applications (accepted)Google Scholar