ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

WRITER-INDEPENDENT OFF-LINE SIGNATURE VERIFICATION

Journal: JOURNAL OF COMPUTER ENGINEERING & TECHNOLOGY (JCET) (Vol.9, No. 4)

Publication Date:

Authors : ; ;

Page : 85-89

Keywords : Center of Excellence for Document Analysis (CEDAR); Artificial Neural network (ANN); Support Vector Machine (SVM); Average Object Area; Entropy; Knearest neighbor (kNN);

Source : Downloadexternal Find it from : Google Scholarexternal

Abstract

In this paper, we address Writer-independent offline signature verification by proposing a new feature extraction algorithm based on no. of objects and entropy of signature image Offline signature verification is a difficult task because dynamic information i.e. temporal information is missing in static image. There is no standard feature extraction method for offline signature recognition as in case of automatic speech recognition like Mel frequency Cestrum Computation (MFCC).Our research presents an intelligent algorithm for offline signature verification using a support vector machine, which is a non-linear classifier. CEDAR (Center of Excellence for Document Analysis) database which have been widely used in signature verification research is used. The database consists of static signature samples taken from 55users.

Last modified: 2018-09-15 15:53:02