{"title":"Investigation of off-line Japanese signature verification using a pattern matching","authors":"K. Ueda","doi":"10.1109/ICDAR.2003.1227800","DOIUrl":null,"url":null,"abstract":"This paper deals with an off-line Japanese signatureverification using a pattern matching method. Anannoying problem encountered in off-line signatureverification using a pattern matching is that variation ofsignature stroke widths and a registered signatureselected from a set of reference samples affect theverification performance. In this paper, we propose amodified pattern matching method, which is independentof stroke widths, and an appropriate selection method ofa registered signature to solve this problem. Thesemethods were experimentally compared with aconventional pattern matching method and a randomselection method of a registered signature. Experimentalresults showed that the proposed methods improve theverification performance.","PeriodicalId":249193,"journal":{"name":"Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings.","volume":"21 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2003-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"29","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings.","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICDAR.2003.1227800","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 29
Abstract
This paper deals with an off-line Japanese signatureverification using a pattern matching method. Anannoying problem encountered in off-line signatureverification using a pattern matching is that variation ofsignature stroke widths and a registered signatureselected from a set of reference samples affect theverification performance. In this paper, we propose amodified pattern matching method, which is independentof stroke widths, and an appropriate selection method ofa registered signature to solve this problem. Thesemethods were experimentally compared with aconventional pattern matching method and a randomselection method of a registered signature. Experimentalresults showed that the proposed methods improve theverification performance.