基于纹理免疫的JPEG预压缩图像降尺度因子检测

doi:10.19678/j.issn.1000-3428.0061303

摘要/Abstract

摘要： 图像编辑工具的普及导致JPEG图像越来越容易被篡改，其中重采样操作方法通过几何变换能够使伪造的图像更加逼真，因此对JPEG图像进行重采样检测至关重要。传统的光谱分析法依据差分图像相邻极值的间隔遵循几何分布，以及在直方图上的峰值呈现周期性，通过峰值分析对下采样因子进行估计，但由于图像纹理统计特征的周期性干扰了直方图的提取，导致检测准确性较低。提出一种用于JPEG预压缩图像降尺度因子检测的纹理免疫块效应分析算法，利用快速导向滤波对图像进行预处理，并去除图像的纹理和噪声。使用Canny算子对滤波后的图像进行边缘检测，以减轻周期性边缘的干扰，对图像进行交叉差分，从而凸显块效应，提高JPEG图像降尺度因子估计的准确性。在此基础上，结合极大似然估计和谱分析得出降尺度因子的估计值，减小估计误差。实验结果表明，该算法能有效削弱图像纹理对重采样估计的影响，具有较强的纹理免疫能力。

关键词: 纹理免疫, JPEG预压缩图像, 快速导向滤波, 降尺度因子估计, 图像重采样检测, JPEG块效应, 交叉差分

Abstract: The popularity of image editing tools makes JPEG images increasingly vulnerable to tampering, and the resampling operation method makes forged images more realistic through geometric transformation.Resamplingof the detection of JPEG images is therefore crucial.The traditional spectral analysis method follows a geometric distribution according to the interval of adjacent extreme values of the difference image and presents periodic peaks on the histogram.Thedownsampling factor is estimated through peak analysis.The periodicity oftheimage texture statistical features interferes with the extraction of the histogram, resulting in low detection accuracy.A texture immune block effect analysis algorithm for the detection of the downscaling factor of the pre-JPEG compressed images is proposed.The image is pre-processed by fast-guided filtering, and the texture and image noise are removed.The Canny operator detects the edge of the filtered image to reduce the interference of periodic edges, and the cross difference highlights the block effect and improves the accuracy of the JPEG image downscaling factor estimation.On this basis, combined with maximum likelihood estimation and spectral analysis, the downscaling factor is estimated to reduce the estimation error. Experimental results show that the algorithm can effectively weaken the influence of image texture on resampling estimation and provide strong texture immunity.

Key words: texture-immune, pre-JPEG compressed image, fast guide filtering, downscaling factor estimation, image resampling detection, JPEG block artifacts, cross-difference

中图分类号:

TP391.41

党良慧, 张玉金, 路东生. 基于纹理免疫的JPEG预压缩图像降尺度因子检测[J]. 计算机工程, 2022, 48(5): 272-280.

DANG Lianghui, ZHANG Yujin, LU Dongsheng. Downscaling Factor Detection of Pre-JPEG Compressed Images Based on Texture-Immune[J]. Computer Engineering, 2022, 48(5): 272-280.

https://www.ecice06.com/CN/Y2022/V48/I5/272

图/表 8

20220806174943

20220806174946

20220806174950

20220806174953

20220806174958

20220806175003

20220806175007

20220806175010

参考文献

[1] MAHDIAN B, SAIC S.Blind authentication using periodic properties of interpolation[J].IEEE Transactions on Information Forensics and Security, 2008, 3(3):529-538.
[2] 周治平, 胡成燕, 黄浩.基于色彩一致性的图像模糊篡改检测[J].计算机工程, 2016, 42(1):237-242. ZHOU Z P, HU C Y, HUANG H.Image blur forgery detection based on color consistency[J].Computer Engineering, 2016, 42(1):237-242.(in Chinese)
[3] KIM D, AHN W, LEE H K.End-to-end anti-forensics network of single and double JPEG detection[J].IEEE Access, 2021, 9:13390-13402.
[4] BIRAJDAR G K, MANKAR V H.Blind method for rescaling detection and rescale factor estimation in digital images using periodic properties of interpolation[J].AEU-International Journal of Electronics and Communications, 2014, 68(7):644-652.
[5] 赵洁, 郭继昌.基于JPEG系数变化率的图像复制粘贴篡改检测[J].浙江大学学报(工学版), 2015, 49(10):1893-1901, 1998. ZHAO J, GUO J C.Image copy-paste forgery detection based on JPEG coefficients change rate[J].Journal of Zhejiang University (Engineering Science), 2015, 49(10):1893-1901, 1998.(in Chinese)
[6] LUO S J, LIU J R, XU W B, et al.Upscaling factor estimation on pre-JPEG compressed images based on difference histogram of spectral peaks[J].Signal Processing:Image Communication, 2021, 94:116223.
[7] 郑二功, 平西建.针对一类JPEG图像伪造的被动盲取证[J].电子与信息学报, 2010, 32(2):394-399. ZHENG E G, PING X J.Passive-blind forensics for a class of JPEG image forgery[J].Journal of Electronics & Information Technology, 2010, 32(2):394-399.(in Chinese)
[8] ZHANG Q, LU W, HUANG T, et al.On the robustness of JPEG post-compression to resampling factor estimation[J].Signal Processing, 2020, 168:107371.
[9] YAO H, MAO F Y, QIN C, et al.Dual-JPEG-image reversible data hiding[J].Information Sciences, 2021, 563:130-149.
[10] KIRCHNER M, GLOE T.On resampling detection in re-compressed images[C]//Proceedings of IEEE International Workshop on Information Forensics and Security.Washington D.C., USA:IEEE Press, 2009:21-25.
[11] BIANCHI T, PIVA A.Reverse engineering of double JPEG compression in the presence of image resizing[C]//Proceedings of IEEE International Workshop on Information Forensics and Security.Washington D.C., USA:IEEE Press, 2012:127-132.
[12] GALLAGHER A C.Detection of linear and cubic interpolation in JPEG compressed images[C]//Proceedings of the 2nd Canadian Conference on Computer and Robot Vision.Washington D.C., USA:IEEE Press, 2005:65-72.
[13] VAZQUEZ-PADÍN D, MOSQUERA C, PÉREZ-GONZÁLEZ F.Two-dimensional statistical test for the presence of almost cyclostationarity on images[C]//Proceedings of IEEE International Conference on Image Processing.Washington D.C., USA:IEEE Press, 2010:1745-1748.
[14] LIU X J, LU W, ZHANG Q, et al.Downscaling factor estimation on pre-JPEG compressed images[J].IEEE Transactions on Circuits and Systems for Video Technology, 2020, 30(3):618-631.
[15] 刘丽, 匡纲要.图像纹理特征提取方法综述[J].中国图象图形学报, 2009, 14(4):622-635. LIU L, KUANG G Y.Overview of image textural feature extraction methods[J].Journal of Image and Graphics, 2009, 14(4):622-635.(in Chinese)
[16] CHEN C L, NI J Q, SHEN Z Y, et al.Blind forensics of successive geometric transformations in digital images using spectral method:theory and applications[J].IEEE Transactions on Image Processing, 2017, 26(6):2811-2824.
[17] KAO Y T, LIN H J, WANG C W, et al.Effective detection for linear up-sampling by a factor of fraction[J].IEEE Transactions on Image Processing, 2012, 21(8):3443-3453.
[18] CHEN Z P, ZHAO Y, NI R R.Detection of operation chain:JPEG-resampling-JPEG[J].Signal Processing:Image Communication, 2017, 57:8-20.
[19] HE K M, SUN J, TANG X O.Guided image filtering[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(6):1397-1409.
[20] KUMAR A, RAHEJA S.Edge detection using guided image filtering and enhanced ant colony optimization[J].Procedia Computer Science, 2020, 173:8-17.
[21] NIKOUKHAH T, GROMPONE VON GIOI R, COLOM M, et al.Automatic JPEG grid detection with controlled false alarms, and its image forensic applications[C]//Proceedings of IEEE Conference on Multimedia Information Processing and Retrieval.Washington D.C., USA:IEEE Press, 2018:378-383.
[22] LIN W S, TJOA S K, ZHAO H V, et al.Digital image source coder forensics via intrinsic fingerprints[J].IEEE Transactions on Information Forensics and Security, 2009, 4(3):460-475.
[23] LI W H, YUAN Y, YU N H.Passive detection of doctored JPEG image via block artifact grid extraction[J].Signal Processing, 2009, 89(9):1821-1829.
[24] CHEN Y L, HSU C T.Image tampering detection by blocking periodicity analysis in JPEG compressed images[C]//Proceedings of the 10th Workshop on Multimedia Signal Processing.Washington D.C., USA:IEEE Press, 2008:803-808.
[25] LI J W, YANG F, LU W, et al.Keypoint-based copy-move detection scheme by adopting MSCRs and improved feature matching[J].Multimedia Tools and Applications, 2017, 76(20):20483-20497.
[26] CHEN L K, LU W, NI J Q, et al.Region duplication detection based on Harris corner points and step sector statistics[J].Journal of Visual Communication and Image Representation, 2013, 24(3):244-254.
[27] ZHANG Q B, LU W, WENG J.Joint image splicing detection in DCT and Contourlet transform domain[J].Journal of Visual Communication and Image Representation, 2016, 40:449-458.
[28] XUE F, YE Z Y, LU W, et al.MSE period based estimation of first quantization step in double compressed JPEG images[J].Signal Processing:Image Communication, 2017, 57:76-83.

选择文件类型/文献管理软件名称

选择包含的内容