基于通道间相关性的图像重着色检测

doi:10.11959/j.issn.2096-109x.2022057

网络与信息安全学报 ›› 2022, Vol. 8 ›› Issue (5): 167-178.doi: 10.11959/j.issn.2096-109x.2022057

基于通道间相关性的图像重着色检测

陈诺¹, 祁树仁¹, 张玉书¹, 薛明富¹, 花忠云²

¹ 南京航空航天大学计算机科学与技术学院，江苏南京 211106
² 哈尔滨工业大学（深圳）计算机科学与技术学院，广东深圳 518055

修回日期:2022-07-01 出版日期:2022-10-15 发布日期:2022-10-01
作者简介:陈诺（1996- ），男，安徽六安人，南京航空航天大学硕士生，主要研究方向为数字图像内容取证
祁树仁（1994- ），男，辽宁朝阳人，南京航空航天大学博士生，主要研究方向为视觉表征、稳健模式识别和媒体内容安全
张玉书（1987- ），男，甘肃庆阳人，南京航空航天大学教授、博士生导师，主要研究方向为多媒体安全与人工智能、区块链与物联网安全、云计算与大数据安全
薛明富（1986- ），男，江苏南京人，南京航空航天大学副教授，主要研究方向为人工智能安全、硬件安全、硬件木马检测和深度学习模型的版权保护
花忠云（1989- ），男，湖南湘西州人，哈尔滨工业大学（深圳）副教授、博士生导师，主要研究方向为混沌理论及应用、多媒体安全、信息隐藏和图像处理
基金资助:
南京航空航天大学研究生科研与实践创新计划项目(xcxjh20211606)

Image recoloring detection based on inter-channel correlation

Nuo CHEN¹, Shuren QI¹, Yushu ZHANG¹, Mingfu XUE¹, Zhongyun HUA²

¹ College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China
² College of Computer Science and Technology, Harbin Institute of Technology, Shenzhen 518055, China

Revised:2022-07-01 Online:2022-10-15 Published:2022-10-01
Supported by:
Graduate Research and Practice Innovation Program Project of Nanjing University of Aeronautics and Astronautics(xcxjh20211606)

摘要/Abstract

摘要：

图像重着色是一种新兴的图像编辑技术，通过篡改像素值达到改变图像颜色风格的目的。随着社交网络和图像编辑技术的快速发展，重着色图像已经严重阻碍了信息传达的真实性。然而，专门为重着色而设计的工作少之又少，现有的重着色检测方法在传统重着色场景下仍有很大提升空间，在应对手工重着色图像时效果不佳。为此，提出了一种基于通道间相关性的重着色图像检测方法，该方法适用于重着色任务中的传统重着色和手工重着色场景。基于相机成像和重着色图像生成方式之间存在显著差异这一现象，提出重着色操作或许会破坏自然图像的通道间相关性这一假设。通过数值分析说明，通道间相关性差异可作为区分重着色图像和自然图像的重要鉴别度量。基于上述先验知识，所提方法通过提取差分图像的一阶微分残差的通道共生矩阵，获得图像的通道间相关性特征集。此外，根据实际情况，假设了3种检测场景，包括训练-测试数据之间匹配、不匹配以及手工重着色场景。实验结果表明，所提方法能够准确识别重着色图像，在假设的3种场景下均优于现有方法，取得了较高的检测精度。除此之外，所提方法对训练数据量的依赖性较小，在训练数据有限的情况下，能实现相当精确的预测结果。

关键词: 图像重着色, 篡改检测, 通道间相关性, 图像取证

Abstract:

Image recoloring is an emerging editing technique that can change the color style of an image by modifying pixel values.With the rapid proliferation of social networks and image editing techniques, recolored images have seriously hampered the authenticity of the communicated information.However, there are few works specifically designed for image recoloring.Existing recoloring detection methods still have much improvement space in conventional recoloring scenarios and are ineffective in dealing with hand-crafted recolored images.For this purpose, a recolored image detection method based on inter-channel correlation was proposed for conventional recoloring and hand-crafted recoloring scenarios.Based on the phenomenon that there were significant disparities between camera imaging and recolored image generation methods, the hypothesis that recoloring operations might destroy the inter-channel correlation of natural images was proposed.The numerical analysis demonstrated that the inter-channel correlation disparities can be used as an important discriminative metric to distinguish between recolored images and natural images.Based on such new prior knowledge, the proposed method obtained the inter-channel correlation feature set of the image.The feature set was extracted from the channel co-occurrence matrix of the first-order differential residuals of the differential image.In addition, three detection scenarios were assumed based on practical situations, including scenarios with matching and mismatching between training-testing data, and scenario with hand-crafted recoloring.Experimental results show that the proposed method can accurately identify recolored images and outperforms existing methods in all three hypothetical scenarios, achieving state-of-the-art detection accuracy.In addition, the proposed method is less dependent on the amount of training data and can achieve fairly accurate prediction results with limited training data.

Key words: image recoloring, forgery detection, inter-channel correlation, image forensics

中图分类号:

TP393

陈诺, 祁树仁, 张玉书, 薛明富, 花忠云. 基于通道间相关性的图像重着色检测[J]. 网络与信息安全学报, 2022, 8(5): 167-178.

Nuo CHEN, Shuren QI, Yushu ZHANG, Mingfu XUE, Zhongyun HUA. Image recoloring detection based on inter-channel correlation[J]. Chinese Journal of Network and Information Security, 2022, 8(5): 167-178.

图/表 15

图1

图2

图3

图4

表1

图5

图6

表2

图7

表3

在测试集D3上的比较结果Table 3 Comparison results on the testing set D3"

方法	准确率	AUC
RecDeNet	83.50%	90.75%
SC-Method	77.00%	84.35%
RF-Method	69.00%	—
DCC-Method	51.50%	50.18%
ICC-Method	$90.50 %$	$94.68 %$

表3

图8

图9

表4

在测试集D4上的比较结果Table 4 Comparison results on the testing set D4"

方法	准确率
RecDeNet	65.00%
SC-Method	70.00%
RF-Method	37.50%
DCC-Method	82.50%
ICC-Method	$85.00 %$

表4

图10

图11

参考文献 46

[1]	NIGHTINGALE S J , WADE K A , WATSON D G . Can people identify original and manipulated photos of real-world scenes[J]. Cognitive Research:Principles and Implications, 2017,2(1): 30.
[2]	VERDOLIVA L . Media forensics and DeepFakes:an overview[J]. IEEE Journal of Selected Topics in Signal Processing, 2020,14(5): 910-932.
[3]	REINHARD E , ADHIKHMIN M , GOOCH B ,et al. Color transfer between images[J]. IEEE Computer Graphics and Applications, 2001,21(5): 34-41.
[4]	PITIE F , KOKARAM A C , DAHYOT R . N-dimensional probability density function transfer and its application to color transfer[C]// Proceedings of Tenth IEEE International Conference on Computer Vision (ICCV'05). 2005: 1434-1439.
[5]	TAI Y W , JIA J , TANG C K . Local color transfer via probabilistic segmentation by expectation-maximization[C]// Proceedings of 2005 IEEE Conference on Computer Vision and Pattern Recognition. 2005: 747-754.
[6]	BEIGPOUR S , VAN DE WEIJER J . Object recoloring based on intrinsic image estimation[C]// Proceedings of 2011 International Conference on Computer Vision. 2011: 327-334.
[7]	PITIé F , KOKARAM A C , DAHYOT R . Automated colour grading using colour distribution transfer[J]. Computer Vision and Image Understanding, 2007,107(1/2): 123-137.
[8]	PITIé F , KOKARAM A , . The linear monge-kantorovitch linear colour mapping for example-based colour transfer[C]// European Conference on Visual Media Production. 2007: 1-9.
[9]	GROGAN M , PRASAD M , DAHYOT R . L2 registration for colour transfer[C]// Proceedings of 2015 23rd European Signal Processing Conference (EUSIPCO). 2015: 1-5.
[10]	LEVIN A , LISCHINSKI D , WEISS Y . Colorization using optimization[M]// ACM Siggraph, 2004: 689-694.
[11]	QU Y , WONG T T , HENG P A . Manga colorization[J]. ACM Transactions on Graphics (TOG), 2006,25(3): 1214-1220.
[12]	AN X B , PELLACINI F . AppProp:all-pairs appearance-space edit propagation[C]// Proceedings of ACM SIGGRAPH 2008. 2008: 1-9.
[13]	XU K , LI Y , JU T ,et al. Efficient affinity-based edit propagation using KD tree[J]. ACM Transactions on Graphics (TOG), 2009,28(5): 1-6.
[14]	CHEN X W , ZOU D Q , LI J W ,et al. Sparse dictionary learning for edit propagation of high-resolution images[C]// Proceedings of 2014 IEEE Conference on Computer Vision and Pattern Recognition. 2014: 2854-2861.
[15]	O'DONOVAN P , AGARWALA A , HERTZMANN A . Color compatibility from large datasets[M]// ACM Siggraph, 2011: 1-12.
[16]	LIN S , RITCHIE D , FISHER M ,et al. Probabilistic color-by-numbers:Suggesting pattern colorizations using factor graphs[J]. ACM Transactions on Graphics (TOG), 2013,32(4): 1-12.
[17]	CHANG H , FRIED O , LIU Y ,et al. Palette-based photo recoloring[J]. ACM Transactions on Graphics (TOG), 2015,34(4): 1-11.
[18]	TSAI Y H , SHEN X H , LIN Z ,et al. Sky is not the limit:semantic-aware sky replacement[J]. ACM Transactions on Graphics, 2016,35(4): 1-11.
[19]	LEE J Y , SON H , LEE G ,et al. Deep color transfer using histogram analogy[J]. The Visual Computer, 2020,36(10): 2129-2143.
[20]	HE M , LIAO J , CHEN D ,et al. Progressive color transfer with dense semantic correspondences[J]. ACM Transactions on Graphics (TOG), 2019,38(2): 1-18.
[21]	ENDO Y , IIZUKA S , KANAMORI Y ,et al. DeepProp:extracting deep features from a single image for edit propagation[J]. Computer Graphics Forum, 2016,35(2): 189-201.
[22]	ZHANG R , ZHU J Y , ISOLA P ,et al. Real-time user-guided image colorization with learned deep priors[J]. arXiv preprint arXiv:1705.02999, 2017.
[23]	CHO J , YUN S , LEE K ,et al. PaletteNet:image recolorization with given color palette[C]// Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops. 2017: 1058-1066.
[24]	AFIFI M , PRICE B L , COHEN S ,et al. Image recoloring based on object color distributions[C]// Eurographics (Short Papers). 2019: 33-36.
[25]	LUAN F , PARIS S , SHECHTMAN E ,et al. Deep photo style transfer[C]// Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. 2017: 4990-4998.
[26]	LI Y J , LIU M Y , LI X T ,et al. A closed-form solution to photorealistic image stylization[C]// Proceedings of the European Conference on Computer Vision. 2018: 453-468.
[27]	YOO J , UH Y , CHUN S ,et al. Photorealistic style transfer via wavelet transforms[C]// Proceedings of 2019 IEEE/CVF International Conference on Computer Vision (ICCV). 2019: 9035-9044.
[28]	BIRAJDAR G K , MANKAR V H . Digital image forgery detection using passive techniques:a survey[J]. Digital Investigation, 2013,10(3): 226-245.
[29]	吴金海, 林福宗 . 基于数字水印的图像认证技术[J]. 计算机学报, 2004,27(9): 1153-1161.
	WU J H , LIN F Z . Image authentication based on digital watermarking[J]. Chinese Journal of Computers, 2004,27(9): 1153-1161.
[30]	YERUSHALMY I , HEL-OR H , . Digital image forgery detection based on lens and sensor aberration[J]. International Journal of Computer Vision, 2011,92(1): 71-91.
[31]	POPESCU A C , FARID H . Statistical tools for digital forensics[C]// International Workshop on Information Hiding. 2004: 128-147.
[32]	LYU S W , PAN X Y , ZHANG X . Exposing region splicing forgeries with blind local noise estimation[J]. International Journal of Computer Vision, 2014,110(2): 202-221.
[33]	MAHDIAN B , SAIC S . Using noise inconsistencies for blind image forensics[J]. Image and Vision Computing, 2009,27(10): 1497-1503.
[34]	FRIDRICH A J , SOUKAL B D , LUKá? A J , . Detection of copy-move forgery in digital images[C]// Digital Forensic Research Workshop. 2003.
[35]	YAN Y Y , REN W Q , CAO X C . Recolored image detection via a deep discriminative model[J]. IEEE Transactions on Information Forensics and Security, 2019,14(1): 5-17.
[36]	ZHANG Y S , CHEN N , QI S R ,et al. Detecting recolored image by spatial correlation[J]. arXiv preprint arXiv:2204.10973, 2022.
[37]	HO J S , AU O C , ZHOU J T ,et al. Inter-channel demosaicking traces for digital image forensics[C]// Proceedings of 2010 IEEE International Conference on Multimedia and Expo. 2010: 1475-1480.
[38]	VAN DE WEIJER J , GEVERS T , GIJSENIJ A . Edge-based color constancy[J]. IEEE Transactions on Image Processing:a Publication of the IEEE Signal Processing Society, 2007,16(9): 2207-2214.
[39]	CARVALHO T , FARIA F A , PEDRINI H ,et al. Illuminant-based transformed spaces for image forensics[J]. IEEE Transactions on Information Forensics and Security, 2016,11(4): 720-733.
[40]	SIMONYAN K , ZISSERMAN A . Very deep convolutional networks for large-scale image recognition[J]. arXiv preprint arXiv:1409.1556, 2014.
[41]	GUNTURK B K , ALTUNBASAK Y , MERSEREAU R M . Color plane interpolation using alternating projections[J]. IEEE Transactions on Image Processing:a Publication of the IEEE Signal Processing Society, 2002,11(9): 997-1013.
[42]	RUSSAKOVSKY O , DENG J , SU H ,et al. ImageNet large scale visual recognition challenge[J]. International Journal of Computer Vision, 2015,115(3): 211-252.
[43]	CHANG C C , LIN C J . LIBSVM:a library for support vector machines[J]. ACM Transactions on Intelligent Systems and Technology, 2011,2(3): 27.
[44]	GOLJAN M , FRIDRICH J , COGRANNE R . Rich model for Steganalysis of color images[C]// Proceedings of 2014 IEEE International Workshop on Information Forensics and Security. 2014: 185-190.
[45]	LI H D , LI B , TAN S Q ,et al. Identification of deep network generated images using disparities in color components[J]. Signal Processing, 2020,174:107616.
[46]	LIN T Y , MAIRE M , BELONGIE S ,et al. Microsoft coco:common objects in context[C]// European Conference on Computer Vision. 2014: 740-755.

数据集	用途	重着色方法	重着色图像数目	自然图像来源	自然图像数目
D₁	训练	文献[3]	18 000	ImageNet^[42]	18 000
		文献[7]	18 000	ImageNet^[42]	18 000
D₂	测试	文献[3]	60	ImageNet^[42]	60
		文献[7]	60	ImageNet^[42]	60
D₃	测试	文献[3,6-9,12,17]	100	下载自网络	100
D₄	测试	手工制作	80	—	—

测试集	准确率	AUC
D₂	93.75%	96.88%
D₃	90.50%	94.68%
D₄	85.00%	—

基于通道间相关性的图像重着色检测

Image recoloring detection based on inter-channel correlation

在线阅读

pdf下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 15

参考文献 46

相关文章 9

Metrics

推荐阅读 0

[1]	章登勇, 文凰, 李峰, 曹鹏, 向凌云, 杨高波, 丁湘陵. 基于双分支网络的图像修复取证方法[J]. 网络与信息安全学报, 2022, 8(6): 110-122.
[2]	林佳滢, 周文柏, 张卫明, 俞能海. 空域频域相结合的唇型篡改检测方法[J]. 网络与信息安全学报, 2022, 8(6): 146-155.
[3]	乔通, 姚宏伟, 潘彬民, 徐明, 陈艳利. 基于深度学习的数字图像取证技术研究进展[J]. 网络与信息安全学报, 2021, 7(5): 13-28.
[4]	肖辉,翁彬,黄添强,普菡,黄则辉. 融合多特征的视频帧间篡改检测算法[J]. 网络与信息安全学报, 2020, 6(1): 84-93.
[5]	张雪莉,黄添强,林晶,黄维. 基于非负张量分解的视频篡改检测方法[J]. 网络与信息安全学报, 2017, 3(6): 42-49.
[6]	黄惠芬,王志红,常玉红. 数学形态学在数字伪造模糊检测中的应用[J]. 网络与信息安全学报, 2017, 3(4): 20-25.
[7]	姚晔,胡伟通,任一支. 视频目标移除篡改的被动取证[J]. 网络与信息安全学报, 2017, 3(2): 46-52.
[8]	黄维,黄添强,张雪莉,肖辉. 基于块效应网格偏移的重获取JPEG图像篡改检测[J]. 网络与信息安全学报, 2017, 3(12): 38-44.
[9]	冯丙文,翁健,卢伟. 基于隐私保护的数字图像取证外包技术框架研究[J]. 网络与信息安全学报, 2016, 2(8): 23-31.