Share Email Print

Proceedings Paper

Off-line signature verification with generated training samples
Author(s): Bin Fang; C. H. Leung; Yuan Y. Tang
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

It is often difficult to obtain sufficient signature samples to train up a signature verification system. With insufficient training samples, the estimation of the statistical parameters such as the mean feature vector and covariance matrix becomes unreliable and system performance would be degraded. In this paper, a two-dimensional elastic matching method to generate additional samples is proposed to expand the limited training set so that a better estimate of the statistical variations can be obtained. The method differs from existing ones in that it is more suitable for the generation of signature samples. Besides this, a set of peripheral features, which is useful in describing both the internal and external structures of signatures, is employed to represent the signatures in the verification process. Results showed that verification system with the additional samples could reduce the error rate from 15.6% to 11.4%. It also outperformed another existing method which estimates the class covariance matrix through optimization techniques. Results also demonstrated that the peripheral features are useful for signature verification.

Paper Details

Date Published: 16 September 2002
PDF: 10 pages
Proc. SPIE 4929, Optical Information Processing Technology, (16 September 2002); doi: 10.1117/12.483241
Show Author Affiliations
Bin Fang, The Univ. of Hong Kong (Singapore)
C. H. Leung, The Univ. of Hong Kong (China)
Yuan Y. Tang, Hong Kong Baptist Univ. (Hong Kong)

Published in SPIE Proceedings Vol. 4929:
Optical Information Processing Technology
Guoguang Mu; Francis T. S. Yu; Suganda Jutamulia, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?