{"title":"基于加密原语的感知隐私的图像认证","authors":"Haixia Chen;Xinyi Huang;Wei Wu;Yi Mu","doi":"10.1093/comjnl/bxaa127","DOIUrl":null,"url":null,"abstract":"Image authentication is the process of verifying image origin, integrity and authenticity. In many situations, image authentication should allow reasonable image editing, which does not introduce any wrong information against the original one. While it has been studied both extensively and intensively with considerable efforts, there is no satisfactory method supporting region extraction. This paper presents a solution to address the issue of privacy protection in authenticated images. Our scheme allows anyone to extract sub-image blocks from an original image (authenticated by the image producer) and generate a proof tag to prove the credibility of the extracted image blocks. The process of proof tag generation does not require any interaction with the image producer. In addition, the image producer is able to define must-be-preserved image blocks (e.g. producer logo) during the extraction. We define the security property for the authenticated sub-images and give a generic design with two core primitives: an ordinary digital signature scheme and a cryptographic accumulator. The security of our design can be reduced to the underlying cryptographic primitives and its practical performance is demonstrated by a bunch of evaluations. We believe the proposed design, together with other image authentication methods, will further facilitate image relevant services and applications.","PeriodicalId":50641,"journal":{"name":"Computer Journal","volume":"64 8","pages":"1178-1192"},"PeriodicalIF":1.5000,"publicationDate":"2020-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1093/comjnl/bxaa127","citationCount":"1","resultStr":"{\"title\":\"Privacy-Aware Image Authentication from Cryptographic Primitives\",\"authors\":\"Haixia Chen;Xinyi Huang;Wei Wu;Yi Mu\",\"doi\":\"10.1093/comjnl/bxaa127\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Image authentication is the process of verifying image origin, integrity and authenticity. In many situations, image authentication should allow reasonable image editing, which does not introduce any wrong information against the original one. While it has been studied both extensively and intensively with considerable efforts, there is no satisfactory method supporting region extraction. This paper presents a solution to address the issue of privacy protection in authenticated images. Our scheme allows anyone to extract sub-image blocks from an original image (authenticated by the image producer) and generate a proof tag to prove the credibility of the extracted image blocks. The process of proof tag generation does not require any interaction with the image producer. In addition, the image producer is able to define must-be-preserved image blocks (e.g. producer logo) during the extraction. We define the security property for the authenticated sub-images and give a generic design with two core primitives: an ordinary digital signature scheme and a cryptographic accumulator. The security of our design can be reduced to the underlying cryptographic primitives and its practical performance is demonstrated by a bunch of evaluations. We believe the proposed design, together with other image authentication methods, will further facilitate image relevant services and applications.\",\"PeriodicalId\":50641,\"journal\":{\"name\":\"Computer Journal\",\"volume\":\"64 8\",\"pages\":\"1178-1192\"},\"PeriodicalIF\":1.5000,\"publicationDate\":\"2020-10-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://sci-hub-pdf.com/10.1093/comjnl/bxaa127\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Computer Journal\",\"FirstCategoryId\":\"94\",\"ListUrlMain\":\"https://ieeexplore.ieee.org/document/9579288/\",\"RegionNum\":4,\"RegionCategory\":\"计算机科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"COMPUTER SCIENCE, HARDWARE & ARCHITECTURE\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Computer Journal","FirstCategoryId":"94","ListUrlMain":"https://ieeexplore.ieee.org/document/9579288/","RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"COMPUTER SCIENCE, HARDWARE & ARCHITECTURE","Score":null,"Total":0}
Privacy-Aware Image Authentication from Cryptographic Primitives
Image authentication is the process of verifying image origin, integrity and authenticity. In many situations, image authentication should allow reasonable image editing, which does not introduce any wrong information against the original one. While it has been studied both extensively and intensively with considerable efforts, there is no satisfactory method supporting region extraction. This paper presents a solution to address the issue of privacy protection in authenticated images. Our scheme allows anyone to extract sub-image blocks from an original image (authenticated by the image producer) and generate a proof tag to prove the credibility of the extracted image blocks. The process of proof tag generation does not require any interaction with the image producer. In addition, the image producer is able to define must-be-preserved image blocks (e.g. producer logo) during the extraction. We define the security property for the authenticated sub-images and give a generic design with two core primitives: an ordinary digital signature scheme and a cryptographic accumulator. The security of our design can be reduced to the underlying cryptographic primitives and its practical performance is demonstrated by a bunch of evaluations. We believe the proposed design, together with other image authentication methods, will further facilitate image relevant services and applications.
期刊介绍:
The Computer Journal is one of the longest-established journals serving all branches of the academic computer science community. It is currently published in four sections.