重庆邮电大学学报（自然科学版）2024，Vol.36Issue(4)：797-806,10.DOI:10.3979/j.issn.1673-825X.202303310091
新颖的内容自适应图像篡改检测与自恢复算法研究
Novel content-adaptive tampering detection and self-recovery algorithm study
摘要
Abstract
In order to address the problem that embedding a large number of watermarks(authentication watermark and ref-erence watermark)in the original image can easily cause image distortion,we propose to embed a small amount of reference watermark.To reduce the image distortion,it is proposed to embed a small amount of reference watermark information in the edges,contours,and other regions of the original image.In recovering the tampered region,it is proposed to transform the tampered block matching problem into a minimization problem of non-local self-similar image blocks using an image restora-tion method,and then replicate the similar information of the attack image itself to recover the tampered region.The experi-mental results show that the algorithm obtains the mean PSNR of the embedded watermarked image over 60 dB and the mean PSNR of the recovered image over 40 dB at a watermark capacity of 0.015 626.关键词
水印/图像恢复/图像修复算法/篡改检测/自恢复Key words
watermarking/image recovery/image inpainting algorithms/tampering detection/self-recovery分类
信息技术与安全科学引用本文复制引用
李明月,毕秀丽,刘波..新颖的内容自适应图像篡改检测与自恢复算法研究[J].重庆邮电大学学报（自然科学版）,2024,36(4):797-806,10.基金项目
国家自然科学基金项目(62172067,61976031) （62172067,61976031）
国家重点研发计划(2019YFE0110800,2016YFC1000307-3) （2019YFE0110800,2016YFC1000307-3）
国家重大科研仪器研制项目(62027827) （62027827）
重庆市教委科学技术研究项目(KJQN202200635)The National Natural Science Foundation of China(62172067,61976031) （KJQN202200635）
The National Key R&D Program(2019YFE0110800,2016YFC1000307-3) （2019YFE0110800,2016YFC1000307-3）
The National Major Research Instrument Development Project(62027827) （62027827）
The Scientif-ic and Technological Research Program of Chongqing Municipal Education Commission(KJQN202200635) （KJQN202200635）
