On-line signature verification based on Gabor features

Abstract

In this paper, the local features obtained from the signature has been used to create the texture image based on the spatio-temporal correlation matrix. The Gabor wavelets exhibit desirable characteristics of spatial locality and orientation selectivity, and are optimally localized in the space and frequency domains. So it has been chosen to extract the Gabor features of the image. Signature is a small sample point problem and Support vector machine (SVM) is especially fit for two classes' problem of small sample. The Gabor features vector has been imported to the SVM classifier and the classification result is obtained. The methods proposed in this paper are validated on the SVC 2004 database and inspiring results are obtained.