Improvement of Offline Signature Verification Method Using Discrete Wavelet Transform

摘要

Signature verification is a widely used personal authentication and verification scheme. To check the legality of different documents such as cheque, visa application, different academic certificates, and attendance checking, it is required to have an automatic offline signature verification system. To solve the abovementioned problem, in this paper we have presented a new transform domain-based offline signature verification scheme. At the first step, spatial features are extracted using the statistical analysis. Then discrete wavelet transform has been used to extract the features in the transform domain. Both the features extracted from the transform domain and spatial domain are combined and fed to the classifier. Different types of classifiers have been used to obtain the recognition rate. The proposed method has been applied on MCYT database and provides the maximum accuracy of 94%.