Copy-Move Forgery Detection Based on Automatic Threshold Estimation

Int. J. Sociotechnology Knowl. Dev. Pub Date : 2020-01-01 DOI:10.4018/ijskd.2020010101

A. Hegazi, A. Taha, M. Selim

{"title":"Copy-Move Forgery Detection Based on Automatic Threshold Estimation","authors":"A. Hegazi, A. Taha, M. Selim","doi":"10.4018/ijskd.2020010101","DOIUrl":null,"url":null,"abstract":"Recently,usersandnewsfollowersacrosswebsitesfacemanyfabricatedimages.Moreover,itgoes farbeyondthattothepointofdefamingorimprisoningaperson.Hence,imageauthenticationhas becomeasignificantissue.Oneofthemostcommontamperingtechniquesiscopy-move.Keypointbasedmethodsareconsideredasaneffectivemethodfordetectingcopy-moveforgeries. Insuch methods,thefeatureextractionprocessisfollowedbyapplyingaclusteringtechniquetogroupspatially closekeypoints.Mostclusteringtechniqueshighlydependontheexistenceofaspecificthreshold toterminatetheclustering.Determinationofthemostsuitablethresholdrequiresahugeamountof experiments.Inthisarticle,acopy-moveforgerydetectionmethodisproposed.Theproposedmethod isbasedonautomaticestimationoftheclusteringthreshold.Thecutoffthresholdofhierarchical clusteringisestimatedautomaticallybasedonclusteringevaluationmeasures.Experimentalresults testedonvariousdatasetsshowthattheproposedmethodoutperformsotherrelevantstate-of-the-art methods. KEywoRDS Clustering Evaluation Measures, Copy-Move Detection, Image Forensics, Keypoint-Based Methods, MultipleCopied Matching","PeriodicalId":13656,"journal":{"name":"Int. J. Sociotechnology Knowl. Dev.","volume":"138 1","pages":"1-23"},"PeriodicalIF":0.0000,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Int. J. Sociotechnology Knowl. Dev.","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.4018/ijskd.2020010101","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 6

Abstract

Recently,usersandnewsfollowersacrosswebsitesfacemanyfabricatedimages.Moreover,itgoes farbeyondthattothepointofdefamingorimprisoningaperson.Hence,imageauthenticationhas becomeasignificantissue.Oneofthemostcommontamperingtechniquesiscopy-move.Keypointbasedmethodsareconsideredasaneffectivemethodfordetectingcopy-moveforgeries. Insuch methods,thefeatureextractionprocessisfollowedbyapplyingaclusteringtechniquetogroupspatially closekeypoints.Mostclusteringtechniqueshighlydependontheexistenceofaspecificthreshold toterminatetheclustering.Determinationofthemostsuitablethresholdrequiresahugeamountof experiments.Inthisarticle,acopy-moveforgerydetectionmethodisproposed.Theproposedmethod isbasedonautomaticestimationoftheclusteringthreshold.Thecutoffthresholdofhierarchical clusteringisestimatedautomaticallybasedonclusteringevaluationmeasures.Experimentalresults testedonvariousdatasetsshowthattheproposedmethodoutperformsotherrelevantstate-of-the-art methods. KEywoRDS Clustering Evaluation Measures, Copy-Move Detection, Image Forensics, Keypoint-Based Methods, MultipleCopied Matching

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

基于自动阈值估计的复制移动伪造检测

最近，usersandnewsfollowersacrosswebsitesfacemanyfabricatedimages。Moreover，itgoes farbeyondthattothepointofdefamingorimprisoningaperson。Hence，imageauthenticationhas becomeasignificantissue.Oneofthemostcommontamperingtechniquesiscopy-move.Keypointbasedmethodsareconsideredasaneffectivemethodfordetectingcopy-moveforgeries。Insuch方法，thefeatureextractionprocessisfollowedbyapplyingaclusteringtechniquetogroupspatially closekeypoints。Mostclusteringtechniqueshighlydependontheexistenceofaspecificthreshold toterminatetheclustering。Determinationofthemostsuitablethresholdrequiresahugeamountof experiments.Inthisarticle，acopy-moveforgerydetectionmethodisproposed。Theproposedmethod isbasedonautomaticestimationoftheclusteringthreshold。Thecutoffthresholdofhierarchical clusteringisestimatedautomaticallybasedonclusteringevaluationmeasures。Experimentalresults testedonvariousdatasetsshowthattheproposedmethodoutperformsotherrelevantstate-of-the-art方法。关键词聚类评价测度，复制移动检测，图像取证，基于关键点的方法，多拷贝匹配

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文去求助

来源期刊

Int. J. Sociotechnology Knowl. Dev.

自引率

0.00%

发文量