网络与信息安全学报 ›› 2022, Vol. 8 ›› Issue (6): 110-122.doi: 10.11959/j.issn.2096-109x.2022084

• 学术论文 • 上一篇    下一篇

基于双分支网络的图像修复取证方法

章登勇1,2, 文凰1,2, 李峰1,2, 曹鹏1,2, 向凌云1,2, 杨高波3, 丁湘陵4   

  1. 1 长沙理工大学湖南省综合交通运输大数据智能处理重点实验室,湖南 长沙 410114
    2 长沙理工大学计算机与通信工程学院,湖南 长沙 410114
    3 湖南大学信息科学与工程学院, 湖南 长沙 410082
    4 湖南科技大学计算机科学与工程学院,湖南 湘潭 411004
  • 修回日期:2022-08-17 出版日期:2022-12-15 发布日期:2023-01-16
  • 作者简介:章登勇(1980- ),男,江西南昌人,长沙理工大学副教授,主要研究方向为多媒体信息安全、图像处理与模式识别
    文凰(1996- ),女,湖南望城人,长沙理工大学硕士生,主要研究方向为多媒体信息安全
    李峰(1964- ),男,湖南常德人,长沙理工大学教授,主要研究方向为多媒体信息安全、图像处理与模式识别
    曹鹏(1982- ),男,湖南常德人,长沙理工大学讲师,主要研究方向为图像处理与模式识别、图像取证
    向凌云(1983- ),女,湖南娄底人,长沙理工大学副教授,主要研究方向为信息安全、信息隐藏与数字水印、隐写分析、自然语言处理、模式识别与机器学习
    杨高波(1973- ),湖南岳阳人,湖南大学教授、博士生导师,主要研究方向为图像/视频信息安全、多媒体通信、纹理压缩
    丁湘陵(1981- ),男,湖南株洲人,博士,湖南科技大学副教授,主要研究方向为多媒体内容安全、图像/视频处理、密码分析
  • 基金资助:
    国家自然科学基金(62172059);国家自然科学基金(61972057);国家自然科学基金(62072055);湖南省自然科学基金(2020JJ4626);湖南省自然科学基金(2020JJ4029);湖南省教育厅优秀青年项目(19B004)

Image inpainting forensics method based on dual branch network

Dengyong ZHANG1,2, Huang WEN1,2, Feng LI1,2, Peng CAO1,2, Lingyun XIANG1,2, Gaobo YANG3, Xiangling DING4   

  1. 1 Hunan Provincial Key Laboratory of Intelligent Processing of Big Data on Transportation, Changsha University of Science and Technology, Changsha 410114, China
    2 School of Computer and Communication Engineering, Changsha University of Science and Technology, Changsha, 410114, China
    3 School of Information Science and Engineering, Hunan University, Changsha 410082, China
    4 School of Computer Science and Engineering, Hunan University of Science and Technology, Xiangtan 411004, China
  • Revised:2022-08-17 Online:2022-12-15 Published:2023-01-16
  • Supported by:
    The National Natural Science Foundation of China(62172059);The National Natural Science Foundation of China(61972057);The National Natural Science Foundation of China(62072055);Natural Science Foundation of Hunan Province(2020JJ4626);Natural Science Foundation of Hunan Province(2020JJ4029);Scientific Research Fund of Hunan Provincial Education Department of China(19B004)

摘要:

图像修复是一项利用图像已知区域的信息来修复图像中缺失或损坏区域的技术。人们借助以此为基础的图像编辑软件无须任何专业基础就可以轻松地编辑和修改数字图像内容,一旦图像修复技术被用于恶意移除图像的内容,会给真实的图像带来信任危机。目前图像修复取证的研究只能有效地检测某一种类型的图像修复。针对这一问题,提出了一种基于双分支网络的图像修复被动取证方法。双分支中的高通滤波卷积网络先使用一组高通滤波器来削弱图像中的低频分量,然后使用4个残差块提取特征,再进行两次4倍上采样的转置卷积对特征图进行放大,此后使用一个 5×5 的卷积来减弱转置卷积带来的棋盘伪影,生成图像高频分量上的鉴别特征图。双分支中的双注意力特征融合分支先使用预处理模块为图像增添局部二值模式特征图。然后使用双注意力卷积块自适应地集成图像局部特征和全局依赖,捕获图像修复区域和原始区域在内容及纹理上的差异,再对双注意力卷积块提取的特征进行融合。最后对特征图进行相同的上采样,生成图像内容和纹理上的鉴别特征图。实验结果表明该方法在检测移除对象的修复区域上,针对样本块修复方法上检测的F1分数较排名第二的方法提高了2.05%,交并比上提高了3.53%;针对深度学习修复方法上检测的F1分数较排名第二的方法提高了1.06%,交并比提高了1.22%。对结果进行可视化可以看出,在检测修复区域上能够准确地定位移除对象的边缘。

关键词: 图像取证, 图像修复检测, 深度学习, 注意力机制

Abstract:

Image inpainting is a technique that uses information from known areas of an image to repair missing or damaged areas of the image.Image editing software based on it has made it easy to edit and modify the content of digital images without any specialized foundation.When image inpainting techniques are used to maliciously remove the content of an image, it will cause confidence crisis on the real image.Current researches in image inpainting forensics can only effectively detect a certain type of image inpainting.To address this problem, a passive forensic method for image inpainting was proposed, which is based on a two-branch network.The high-pass filtered convolutional network in the dual branch first used a set of high-pass filters to attenuate the low-frequency components in the image.Then features were extracted using four residual blocks, and two transposed convolutions were performed with 4x up-sampling to zoom in on the feature map.And thereafter a 5×5 convolution was used to attenuate the tessellation artifacts from the transposed convolutions to generate a discriminative feature map on the high-frequency components of the image.The dual-attention feature fusion branch in the dual branch first added a local binary pattern feature map to the image using a preprocessing block.Then the dual-attention convolution block was used to adaptively integrate the image’s local features and global dependencies to capture the differences in content and texture between the inpainted and pristine regions of the image.Additionally, the features extracted from the dual-attention convolution block were fused, and the feature maps were up-sampled identically to generate the discriminative image content and texture on the feature maps.The extensive experimental results show the proposed method improved the F1 score by 2.05% and the Intersection over Union(IoU) by 3.53% for the exemplar-based method and by 1.06% and 1.22% for the deep-learning-based method in detecting the inpainted region of the removed object.Visualization of the results shows that the edges of the removed objects can be accurately located on the detected inpainted area.

Key words: image forensics, image forgery detection, deep learning, attention mechanism

中图分类号: 

No Suggested Reading articles found!