面向图像篡改取证的多特征融合U形深度网络

doi:10.19678/j.issn.1000-3428.0061039

摘要/Abstract

摘要： 随着图像篡改工具的智能化发展，图像篡改不再局限于拼接、移除等某一具体的类型，往往包含多种篡改类型及其组合操作，使得图像篡改取证工作更具挑战性。提出一种端到端的多特征融合U形深度网络，利用编解码网络提取篡改区域与真实区域之间的对比度差异、边缘差异等篡改痕迹，并使用富隐写模型卷积层获取伪造图像的噪声分布不规律信息，从而在无预处理的情况下实现可疑区域的检测并分割出高置信度的篡改区域。在此基础上，使用特征提取模块获取融合的篡改特征，在融合定位模块中利用分级监督策略融合不同分辨率提取的篡改特征，以准确定位篡改区域，实现篡改区域检测与像素级的分割。实验结果表明，基于所提网络的图像篡改取证方法在NIST16和CASIA数据库上的F1值分别为0.841和0.605，与基于MFCN、RGB-N、MANTRA-net等网络的图像篡改取证方法相比，有较优的检测性能和较高的实时性，且对JPEG压缩、缩放等处理具有更强的鲁棒性。

关键词: 图像篡改取证, 深度神经网络, 编解码网络, 噪声信息, 富隐写模型

Abstract: With the intelligent development of image tampering tools, the types of image tampering are not limited to a specific type such as splicing and remove, but often includes multiple types of tampering and their combined manipulability, making image forensics more challenging.To solve the complex image tampering detection, an end-to-end Multi-Feature Fusion U-Structure deep network for image forgeries detection (MFF-US net).Using the encoder-decoder architecture to extract the tampering traces, which include the contrast difference, edge and other differences between the manipulated and non-manipulated regions, and using the convolutional layer of the rich models steganalysis to obtain the irregular information of noise distribution on the forgery image, detection of suspicious areas and segmentation of tampered areas with high confidence without pre-processing.On this basis, the feature extraction module is used to obtain the fused tamper features.In the fusion location module, the hierarchical supervision strategy is used to fuse the tamper features extracted from different resolutions, so as to accurately locate the tamper area and realize the tamper area detection and pixel level segmentation.Experiments show the F1 values of the proposed method in the NIST16 and CASIA databases are 0.841 and 0.605, respectively.Compared with the existing MFCN、RGB-N、MANTRA-net mainstream method, this method can not only achieve better detection performance and real-time performance, but also has strong robustness to JPEG compression and post-scaling processing.

Key words: image tampering forensics, deep neural network, encoder-decoder network, noise information, rich steganography model

中图分类号:

TP391.41

路东生, 张玉金, 党良慧. 面向图像篡改取证的多特征融合U形深度网络[J]. 计算机工程, 2022, 48(4): 213-222.

LU Dongsheng, ZHANG Yujin, DANG Lianghui. Multi-Feature Fusion U-structure Deep Network for Image Tempering Forensics[J]. Computer Engineering, 2022, 48(4): 213-222.

http://www.ecice06.com/CN/Y2022/V48/I4/213

图/表 11

20230131201837

20230131201840

20230131201843

20230131201847

20230131201850

20230131201853

20230131201856

20230131201859

20230131201903

20230131201906

20230131201909

参考文献

[1] AMERINI I, URICCHIO T, BALLAN L, et al.Localization of JPEG double compression through multi-domain convolutional neural networks[EB/OL].[2021-02-02] https://arxiv.org/abs/1706.01788.
[2] CHEN J, KANG X, LIU Y, et al.Median filtering forensics based on convolutional neural networks[J].IEEE Signal Processing Letters, 2015, 22(11):1849-1853.
[3] HUH M, LIU A, OWENS A, et al.Fighting fake news:image splice detection via learned self-consistency[C]//Proceedings of European Conference on Computer Vision.Berlin, Germany:Springer, 2018:101-117.
[4] WU Y, ABD-ALMAGEED W, NATARAJAN P.Deep matching and validation network:an end-to-end solution to constrained image splicing localization and detection[C]//Proceedings of the 25th ACM International Conference on Multimedia. New York, USA:ACM Press, 2017:1480-1502.
[5] COZZOLINO D, POGGI G, VERDOLIVA L.Efficient dense-field copy-move forgery detection[J].IEEE Transactions on Information Forensics and Security, 2015, 10(11):2284-2297.
[6] 邢文博, 杜志淳.数字图像复制粘贴篡改取证[J].计算机科学, 2019, 46(S1):380-384, 396. XING W B, DU Z C.Digital image forensics for copy and paste tampering[J].Computer Science, 2019, 46(S1):380-384, 396.(in Chinese)
[7] ZHU X, QIAN Y, ZHAO X, et al.A deep learning approach to patch-based image inpainting forensics[J].Signal Processing:Image Communication, 2018, 67:90-99.
[8] WU Y, ABD-ALMAGEED W, NATARAJAN P.Busternet:detecting copy-move image forgery with source/target localization[C]//Proceedings of European Conference on Computer Vision.Berlin, Germany:Springer, 2018:168-184.
[9] WU Y, ABD-ALMAGEED W, NATARAJAN P.Image copy-move forgery detection via an end-to-end deep neural network[C]//Proceedings of 2018 IEEE Winter Conference on Applications of Computer Vision.Washington D.C., USA:IEEE Press, 2018:1907-1915.
[10] 霍占强, 刘玉洁, 付苗苗, 等.基于卷积神经网络的直线描述方法研究[J].计算机工程, 2021, 47(5):251-259. HUO Z Q, LIU Y J, FU M M, et al.Research on line description method based on convolutional neural network[J].Computer Engineering, 2021, 47(5):251-259.(in Chinese)
[11] BUNK J, BAPPY J H, MOHAMMED T M, et al.Detection and localization of image forgeries using resampling features and deep learning[C]//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops.Washington D.C., USA:IEEE Press, 2017:1881-1889.
[12] BAPPY J H, ROY-CHOWDHURY A K, BUNK J, et al.Exploiting spatial structure for localizing manipulated image regions[C]//Proceedings of IEEE International Conference on Computer Vision.Washington D.C., USA:IEEE Press, 2017:4970-4979.
[13] BONDI L, LAMERI S, GUERA D, et al.Tampering detection and localization through clustering of camera-based CNN features[C]//Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops.Washington D.C., USA:IEEE Press, 2017:1855-1864.
[14] SALLOUM R, REN Y, KUO C C J.Image splicing localization using a multi-task fully convolutional network[J].Journal of Visual Communication and Image Representation, 2018, 51:201-209.
[15] ZHANG Z, ZHANG Y, ZHOU Z, et al.Boundary-based image forgery detection by fast shallow cnn[C]//Proceedings of the 24th International Conference on Pattern Recognition.Washington D.C., USA:IEEE Press, 2018:2658-2663.
[16] COZZOLINO D, VERDOLIVA L.Noiseprint:a CNN-based camera model fingerprint[J].IEEE Transactions on Information Forensics and Security, 2019, 15:144-159.
[17] LI J, LI X, YANG B, et al.Segmentation-based image copy-move forgery detection scheme[J].IEEE transactions on information forensics and security, 2014, 10(3):507-518.
[18] MAHDIAN B, SAIC S.Using noise inconsistencies for blind image forensics[J].Image and Vision Computing, 2009, 27(10):1497-1503.
[19] PAN X, ZHANG X, LYU S.Exposing image splicing with inconsistent local noise variances[C]//Proceedings of 2012 IEEE International Conference on Computational Photography.Washington D.C., USA:IEEE Press, 2012:1-10.
[20] ZENG H, ZHAN Y F, KANG X G, et al.Image splicing localization using PCA-based noise level estimation[J].Multimedia Tools and Applications, 2017, 76(4):4783-4799.
[21] FAN Y, CARRÉ P, FERNANDEZ-MALOIGNE C.Image splicing detection with local illumination estimation[C]//Proceedings of 2015 IEEE International Conference on Image Processing.Washington D.C., USA:IEEE Press, 2015:2940-2944.
[22] 朱新山, 钱永军, 孙彪, 等.基于深度神经网络的图像修复取证算法[J].光学学报, 2018, 38(11):105-113. ZHU X S, QIAN Y J, SUN B, et al.Image inpainting forensics algorithm based on deep neural network[J].Acta Optica Sinica, 2018, 38(11):105-113.(in Chinese)
[23] LIANG Z, YANG G, DING X, et al.An efficient forgery detection algorithm for object removal by exemplar-based image inpainting[J].Journal of Visual Communication and Image Representation, 2015, 30:75-85.
[24] BAPPY J H, SIMONS C, NATARAJ L, et al.Hybrid LSTM and encoder-decoder architecture for detection of image forgeries[J].IEEE Transactions on Image Processing, 2019, 28(7):3286-3300.
[25] MAZAHERI G, MITHUN N C, BAPPY J H, et al.A skip connection architecture for localization of image manipulations[EB/OL].[2021-02-02].https://www.researchgate.net/publication/335463706_A_Skip_Connection_Architecture_for_Localization_of_Image_Manipulations.
[26] MOHAMMED T M, BUNK J, NATARAJ L, et al.Boosting image forgery detection using resampling features and copy-move analysis[J].Electronic Imaging, 2018, 48(7):118-128.
[27] QIN X, ZHANG Z, HUANG C, et al.U2-Net:going deeper with nested U-structure for salient object detection[EB/OL].[2021-02-02].https://arxiv.org/abs/2005.09007.
[28] FRIDRICH J, KODOVSKY J.Rich models for steganalysis of digital images[J].IEEE Transactions on Information Forensics and Security, 2012, 7(3):868-882.
[29] MULTIMODAL INFORMATION GROUP.Open media forensics challenge[EB/OL].[2021-02-02].https://www.nist.gov/itl/iad/mig/open-media-forensics-challenge.
[30] JIE Z, WEI F Z.CASIA tampered image detection evaluation dataset[EB/OL].[2021-02-02].https://www.oalib.com/references/14079387.
[31] TIAN T N, JESSIE H, SHIH F C, Columbia Image splicing detection evaluation dataset[EB/OL].[2021-02-02].https://www.ee.columbia.edu/ln/dvmm/downloads/AuthSpliced DataSet/AuthSplicedDataSet.htm.
[32] KRAWETZ N, SOLUTIONS H F.A picture's worth[J].Hacker Factor Solutions, 2007, 6(2):2-6.
[33] FERRARA P, BIANCHI T, DE ROSA A, et al.Image forgery localization via fine-grained analysis of CFA artifacts[J].IEEE Transactions on Information Forensics and Security, 2012, 7(5):1566-1577.
[34] ZHOU P, HAN X, MORARIU V I, et al.Learning rich features for image manipulation detection[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2018:1053-1061.
[35] WU Y, ABDALMAGEED W, NATARAJAN P.Mantra-net:manipulation tracing network for detection and localization of image forgeries with anomalous features[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2019:9543-9552.

选择文件类型/文献管理软件名称

选择包含的内容