A Novel Video Forgery Detection Model Based on Triangular Polarity Feature Classification

IF 0.6 Q4 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

International Journal of Digital Crime and Forensics Pub Date : 2020-01-01 DOI:10.4018/ijdcf.2020010102

Chee Cheun Huang, Chien Eao Lee, V. Thing

{"title":"A Novel Video Forgery Detection Model Based on Triangular Polarity Feature Classification","authors":"Chee Cheun Huang, Chien Eao Lee, V. Thing","doi":"10.4018/ijdcf.2020010102","DOIUrl":null,"url":null,"abstract":"Videoforgeryhasbeenincreasingovertheyearsduetothewideaccessibilityofsophisticatedvideo editingsoftware.Ahighlyaccurateandautomatedvideoforgerydetectionsystemwillthereforebe vitallyimportantinensuringtheauthenticityofforensicvideoevidences.Thisarticleproposesanovel TriangularPolarityFeatureClassification(TPFC)videoforgerydetectionframeworkforvideoframe insertionanddeletionforgeries.TheTPFCframeworkhashighprecisionandrecallrateswithasimple andthreshold-lessalgorithmdesignedforreal-worldapplications.Systemrobustnessevaluationsbased oncrossvalidationanddifferentdatabaserecordingconditionswerealsoperformedandvalidated. EvaluationontheperformanceoftheTPFCframeworkdemonstratedtheefficacyoftheproposed frameworkbyachievingarecallrateofupto98.26%andprecisionrateofupto95.76%,aswellas highlocalizationaccuracyondetectedforgedvideos.TheTPFCframeworkisfurtherdemonstrated tobecapableofoutperformingothermodernvideoforgerydetectiontechniquesavailabletoday. KeywoRDS Frame Deletion, Frame Insertion, Inter-frame Forgery Detection, Precision Rate, Recall Rate, Video Forensic","PeriodicalId":44650,"journal":{"name":"International Journal of Digital Crime and Forensics","volume":"22 1","pages":"14-34"},"PeriodicalIF":0.6000,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"5","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Digital Crime and Forensics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.4018/ijdcf.2020010102","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS","Score":null,"Total":0}

引用次数: 5

Abstract

Videoforgeryhasbeenincreasingovertheyearsduetothewideaccessibilityofsophisticatedvideo editingsoftware.Ahighlyaccurateandautomatedvideoforgerydetectionsystemwillthereforebe vitallyimportantinensuringtheauthenticityofforensicvideoevidences.Thisarticleproposesanovel TriangularPolarityFeatureClassification(TPFC)videoforgerydetectionframeworkforvideoframe insertionanddeletionforgeries.TheTPFCframeworkhashighprecisionandrecallrateswithasimple andthreshold-lessalgorithmdesignedforreal-worldapplications.Systemrobustnessevaluationsbased oncrossvalidationanddifferentdatabaserecordingconditionswerealsoperformedandvalidated. EvaluationontheperformanceoftheTPFCframeworkdemonstratedtheefficacyoftheproposed frameworkbyachievingarecallrateofupto98.26%andprecisionrateofupto95.76%,aswellas highlocalizationaccuracyondetectedforgedvideos.TheTPFCframeworkisfurtherdemonstrated tobecapableofoutperformingothermodernvideoforgerydetectiontechniquesavailabletoday. KeywoRDS Frame Deletion, Frame Insertion, Inter-frame Forgery Detection, Precision Rate, Recall Rate, Video Forensic

查看原文本刊更多论文

基于三角极性特征分类的视频伪造检测模型

Videoforgeryhasbeenincreasingovertheyearsduetothewideaccessibilityofsophisticatedvideo editingsoftware。Ahighlyaccurateandautomatedvideoforgerydetectionsystemwillthereforebe vitallyimportantinensuringtheauthenticityofforensicvideoevidences。Thisarticleproposesanovel TriangularPolarityFeatureClassification(TPFC)videoforgerydetectionframeworkforvideoframe insertionanddeletionforgeries。TheTPFCframeworkhashighprecisionandrecallrateswithasimple andthreshold-lessalgorithmdesignedforreal-worldapplications。Systemrobustnessevaluationsbased oncrossvalidationanddifferentdatabaserecordingconditionswerealsoperformedandvalidated。EvaluationontheperformanceoftheTPFCframeworkdemonstratedtheefficacyoftheproposed frameworkbyachievingarecallrateofupto98.26%andprecisionrateofupto95.76%，aswellas highlocalizationaccuracyondetectedforgedvideos。TheTPFCframeworkisfurtherdemonstrated tobecapableofoutperformingothermodernvideoforgerydetectiontechniquesavailabletoday。关键词:帧删除，帧插入，帧间伪造检测，准确率，召回率，视频取证

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分钟内获得全文求助全文

来源期刊

International Journal of Digital Crime and Forensics COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS-

CiteScore

2.70

自引率

0.00%

发文量