基于先验分布的抗JPEG压缩的鲁棒图像隐写

doi:10.19678/j.issn.1000-3428.0069224

摘要/Abstract

摘要：

由于网络信道压缩过程不可知, 使得图像在历经社交网络传输之后受到的攻击难以预测, 因此研究抵抗压缩失真信道的隐写算法是一个非常具有挑战的难题。现有的抵抗JPEG压缩的算法在训练时所使用的质量因子是固定的, 而在实际应用中, JPEG压缩的质量因子并不是固定的, 而是依赖于原始图像的特性。因此, 要设计出可以抵抗社交网络中的JPEG失真的方法, 首先需要研究社交网络压缩信道的质量因子分布情况。针对现有抗JPEG压缩的鲁棒隐写算法对社交网络压缩失真脆弱的问题, 研究JPEG压缩质量因子在社交网络传输中的分布特性, 并采用高斯混合模型(GMM)对其进行建模。在隐写模型的训练过程中, 质量因子的选择不再采用传统鲁棒隐写方法中的固定值, 而是从GMM中进行平滑采样。实验证明, 所设计的抗JPEG压缩鲁棒性隐写模型在遭受社交网络JPEG压缩攻击后, 能够显著提升图像的视觉质量, 同时降低信息提取错误率。相较于其他模型, 模型也具有较好的安全性。

关键词: 鲁棒图像隐写, JPEG压缩, 深度神经网络, 社交网络, 先验分布

Abstract:

Because the compression process of the network channel is unknown, attacks on the image after social network transmission are difficult to predict. Therefore, studying the steganography algorithm resistant to the compression distortion of a channel is challenging. However, the quality factor used by existing algorithms that resist JPEG compression during training is fixed, while in practical applications, the quality factor of JPEG compression is not fixed, but depends on the characteristics of the original image. Therefore, to design a method resistant to JPEG distortion in social networks, the quality factor distribution of social network compression channels must be examined. This study aims to design a robust steganography algorithm resistant to JPEG compression distortion during social network compression. This study investigates the distribution characteristics of the JPEG compression quality factor in social network transmission using a Gaussian Mixture Model (GMM). During the training process of the steganography model, the quality factor is no longer fixed as in the traditional steganography method but is smoothly sampled from the GMM. The experimental results show that the proposed steganography model can significantly improve the visual quality of an image and reduce the error rate of information extraction. Moreover, compared with the other models, the steganography security of this model is improved.

Key words: robust image steganography, JPEG compression, deep neural network, social network, prior distribution

刘婷, 董理, 严迪群, 王让定. 基于先验分布的抗JPEG压缩的鲁棒图像隐写[J]. 计算机工程, 2025, 51(11): 235-245.

LIU Ting, DONG Li, YAN Diqun, WANG Rangding. Robust Image Steganography Based on Prior Distribution Resistant to JPEG Compression[J]. Computer Engineering, 2025, 51(11): 235-245.

https://www.ecice06.com/CN/Y2025/V51/I11/235

图/表 13

图1 社交网络质量因子GMM拟合图

Fig.1 GMM fitting diagram of social network quality factor

图2 EM模型流程图

Fig.2 EM model flow chart

图3 基于先验分布的抗JPEG压缩的鲁棒隐写框架

Fig.3 Framework of robust steganography resistant to JPEG compression based on prior distribution

图4 4种模型残差图像对比

Fig.4 Comparison of residual images of four models

图5 隐写图像经过微信压缩攻击后的效果对比

Fig.5 Comparison of the effect of steganographic images after WeChat compression attack

图6 隐写图像经过Facebook压缩攻击后的效果对比

Fig.6 Comparison of the effect of steganographic images after Facebook compression attack

图7 StegExpose隐写分析ROC曲线

Fig.7 ROC curve of StegExpose steganalysis

参考文献 46

1	RIVEST R L. Cryptography[M]//LEEUWEN J. Algorithms and complexity. [S. l. ]: Elsevier, 1990: 717-755.
2	FRIDRICH J, GOLJAN M, SOUKAL D. Searching for the stego-key[M]//DELP E J. Security, steganography, and watermarking of multimedia contents Ⅵ. San Jose, USA: SPIE, 2004: 70.
3	YAMAGUCHI K, BERG T L, ORTIZ L E. Chic or social: visual popularity analysis in online fashion networks[C]//Proceedings of the 22nd ACM International Conference on Multimedia. New York, USA: ACM Press, 2014: 773-776.
4	李政学, 李枝名, 彭德中, 等. 基于特征对比学习和图卷积的社交网络用户分类. 计算机工程, 2024, 50 (4): 258- 266. doi: 10.19678/j.issn.1000-3428.0067327
	LI Z X , LI Z M , PENG D Z , et al. Classification of social network users based on feature comparative learning and graph convolution. Computer Engineering, 2024, 50 (4): 258- 266. doi: 10.19678/j.issn.1000-3428.0067327
5	ABOUT F B. Newsroom[EB/OL]. [2023-12-01]. https://about.fb.com/news/.
6	Zephoria. Digital marketing[EB/OL]. [2023-12-01]. https://zephoria.com/top-15-valuable-facebook-statistics/.
7	朱利妍. 面向有损网络信道图像鲁棒隐写关键问题研究[D]. 郑州: 战略支援部队信息工程大学, 2022.
	ZHU L Y. Research on key issues of robust steganography for lossy network channel images[D]. Zhengzhou: Information Engineering University, 2022. (in Chinese)
8	RA M R, GOVINDAN R, ORTEGA A. P3: toward privacy-preserving photo sharing[C]//Proceedings of the 10th USENIX Symposium on Networked Systems Design and Implementation. [S. l. ]: USENIX, 2013: 515-528.
9	王仰光. 适配社交网络的批量JPEG图像信息隐藏方法研究[D]. 合肥: 中国科学技术大学, 2023.
	WANG Y G. Research on information hiding method of batch JPEG images adapted to social network[D]. Hefei: University of Science and Technology of China, 2023. (in Chinese)
10	SUN W W, ZHOU J T, LYU R, et al. Processing-aware privacy-preserving photo sharing over online social networks[C]//Proceedings of the 24th ACM International Conference on Multimedia. New York, USA: ACM Press, 2016: 581-585.
11	PETITCOLAS F A P , ANDERSON R J , KUHN M G . Information hidin-a survey. Proceedings of the IEEE, 1999, 87 (7): 1062- 1078. doi: 10.1109/5.771065
12	ALMUTIRI M G , OTHMAN M T B . Digital image watermarking based on LSB techniques: a comparative study. International Journal of Computer Applications, 2018, 975, 8887.
13	ZOU D K, SHI Y, SU W, et al. Steganalysis based on Markov model of thresholded prediction-error image[C]//Proceedings of the IEEE International Conference on Multimedia and Expo. Toronto, Canada: IEEE Press, 2006: 1365-1368.
14	YU X Z , CHEN K J , WANG Y F , et al. Robust adaptive steganography based on generalized dither modulation and expanded embedding domain. Signal Processing, 2020, 168, 107343. doi: 10.1016/j.sigpro.2019.107343
15	ZHANG Y , ZHU X D , QIN C , et al. Dither modulation based adaptive steganography resisting JPEG compression and statistic detection. Multimedia Tools and Applications, 2018, 77 (14): 17913- 17935. doi: 10.1007/s11042-017-4506-3
16	SU W K , NI J Q , HU X L , et al. New design paradigm of distortion cost function for efficient JPEG steganography. Signal Processing, 2022, 190, 108319. doi: 10.1016/j.sigpro.2021.108319
17	钱清, 龙永, 蒋忠远, 等. 基于深度强化学习的自适应图像隐写算法. 计算机工程, 2024, 50 (8): 319- 327. doi: 10.19678/j.issn.1000-3428.0068359
	QIAN Q , LONG Y , JIANG Z Y , et al. Adaptive image steganography algorithm based on deep reinforcement learning. Computer Engineering, 2024, 50 (8): 319- 327. doi: 10.19678/j.issn.1000-3428.0068359
18	COGRANNE R , GIBOULOT E , BAS P . Efficient steganography in JPEG images by minimizing performance of optimal detector. IEEE Transactions on Information Forensics and Security, 2022, 17, 1328- 1343. doi: 10.1109/TIFS.2021.3111713
19	GIBOULOT E , BAS P , COGRANNE R . Multivariate side-informed Gaussian embedding minimizing statistical detectability. IEEE Transactions on Information Forensics and Security, 2022, 17, 1841- 1854. doi: 10.1109/TIFS.2022.3173184
20	张静涵, 陈文. 抗JPEG压缩的自适应图像隐写算法. 计算机科学, 2024, 51 (5): 321- 330.
	ZHANG J H , CHEN W . Adaptive image steganography against JPEG compression. Computer Science, 2024, 51 (5): 321- 330.
21	ZHANG J M , ZHAO X F , HE X L . Robust JPEG steganography based on the robustness classifier. EURASIP Journal on Information Security, 2023, 2023 (1): 11. doi: 10.1186/s13635-023-00148-x
22	DUAN X L , LI B , YIN Z X , et al. Robust image steganography against lossy JPEG compression based on embedding domain selection and adaptive error correction. Expert Systems with Applications, 2023, 229, 120416. doi: 10.1016/j.eswa.2023.120416
23	YANG H , XU Y T , LIU X H , et al. PRIS: practical robust invertible network for image steganography. Engineering Applications of Artificial Intelligence, 2024, 133, 108419. doi: 10.1016/j.engappai.2024.108419
24	ZENG K , CHEN K J , ZHANG W M , et al. Robust steganography for high quality images. IEEE Transactions on Circuits and Systems for Video Technology, 2023, 33 (9): 4893- 4906. doi: 10.1109/TCSVT.2023.3250750
25	LI W X , WU S H , LI B , et al. Payload-independent direct cost learning for image steganography. IEEE Transactions on Circuits and Systems for Video Technology, 2024, 34 (3): 1970- 1975. doi: 10.1109/TCSVT.2023.3294291
26	PAN Y F , NI J Q , LIU Q L , et al. Efficient JPEG image steganography using pairwise conditional random field model. Signal Processing, 2024, 221, 109493. doi: 10.1016/j.sigpro.2024.109493
27	ZHU J R, KAPLAN R, JOHNSON J, et al. HiDDeN: hiding data with deep networks[C]//Proceedings of ECCV 2018. Berlin, Germany: Springer, 2018: 682-697.
28	ZHANG K, CUESTA-INFANTE A, XU L, et al. SteganoGAN: high capacity image steganography with GANs[EB/OL]. [2023-12-01]. https://arxiv.org/pdf/1901.03892.
29	ZHANG R , DONG S Q , LIU J Y . Invisible steganography via generative adversarial networks. Multimedia Tools and Applications, 2019, 78 (7): 8559- 8575. doi: 10.1007/s11042-018-6951-z
30	JIA Z Y, FANG H, ZHANG W M. MBRS: enhancing robustness of DNN-based watermarking by mini-batch of real and simulated JPEG compression[C]//Proceedings of the 29th ACM International Conference on Multimedia. New York, USA: ACM Press, 2021: 41-49.
31	YANG J H , SHANG F , LIAO Y , et al. Toward high capacity and robust JPEG steganography based on adversarial training. Security and Communication Networks, 2023, 2023, 3813977.
32	LI G B , LI S , LI M L , et al. Steganography of steganographic networks. Proceedings of the AAAI Conference on Artificial Intelligence, 2023, 37 (4): 5178- 5186. doi: 10.1609/aaai.v37i4.25647
33	QIAO T , XU S W , WANG S , et al. Robust steganography in practical communication: a comparative study. EURASIP Journal on Image and Video Processing, 2023, 2023 (1): 15. doi: 10.1186/s13640-023-00615-y
34	TANCIK M, MILDENHALL B, NG R. StegaStamp: invisible hyperlinks in physical photographs[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle, USA: IEEE Press, 2020: 2114-2123.
35	PEVN AY'G T, BAS P, FRIDRICH J. Steganalysis by subtractive pixel adjacency matrix[C]//Proceedings of the 11th ACM Workshop on Multimedia and Security. New York, USA: ACM Press, 2009: 75-84.
36	FRIDRICH J , KODOVSKY J . Rich models for steganalysis of digital images. IEEE Transactions on Information Forensics and Security, 2012, 7 (3): 868- 882. doi: 10.1109/TIFS.2012.2190402
37	KODOVSKY J , FRIDRICH J , HOLUB V . Ensemble classifiers for steganalysis of digital media. IEEE Transactions on Information Forensics and Security, 2012, 7 (2): 432- 444. doi: 10.1109/TIFS.2011.2175919
38	QIAN Y L, DONG J, WANG W, et al. Deep learning for steganalysis via convolutional neural networks[C]//Proceedings of Conference on Media Watermarking, Security, and Forensics 2015. San Francisco, USA: SPIE, 2015: 94090J.
39	XU G S , WU H Z , SHI Y Q . Structural design of convolutional neural networks for steganalysis. IEEE Signal Processing Letters, 2016, 23 (5): 708- 712. doi: 10.1109/LSP.2016.2548421
40	MANDAL P C. Structural design of convolutional neural network-based steganalysis[C]//Proceedings of the 7th International Conference on Advanced Computing, Networking, and Informatics (ICACNI 2019). Singapore: Springer Singapore, 2021: 39-45.
41	BOROUMAND M , CHEN M , FRIDRICH J . Deep residual network for steganalysis of digital images. IEEE Transactions on Information Forensics and Security, 2019, 14 (5): 1181- 1193. doi: 10.1109/TIFS.2018.2871749
42	YAO Y , HUANG L C , WANG H , et al. Robust adaptive steganography based on adaptive STC-ECC. IEEE Transactions on Multimedia, 2024, 26, 5477- 5489. doi: 10.1109/TMM.2023.3334487
43	MOON T K . The expectation-maximization algorithm. IEEE Signal Processing Magazine, 1996, 13 (6): 47- 60. doi: 10.1109/79.543975
44	DENG J, DONG W, SOCHER R, et al. ImageNet: a large-scale hierarchical image database[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Miami, USA: IEEE Press, 2009: 248-255.
45	KINGA D A. A method for stochastic optimization[C]//Proceedings of International Conference on Learning Representations. San Dego, USA: ICLR, 2015, 7.
46	BOEHM B. StegExpose-a tool for detecting LSB steganography[EB/OL]. [2023-12-01]. https://doi.org/10.48550/arxiv.1410.6656.

[1]	田青, 王斌, 周子枭. 无监督行人重识别研究综述[J]. 计算机工程, 2025, 51(7): 12-30.
[2]	郭佩林, 张德, 王怀秀. 基于特征可视化探究跳跃连接结构对深度神经网络特征提取的影响[J]. 计算机工程, 2025, 51(4): 149-157.
[3]	赵宏, 宋馥荣, 李文改. 基于SE-AdvGAN的图像对抗样本生成方法研究[J]. 计算机工程, 2025, 51(2): 300-311.
[4]	黄舒怡, 谭光. 基于分区的高效视频目标检测[J]. 计算机工程, 2025, 51(2): 65-77.
[5]	路悦, 周翔宇, 张世周, 梁国强, 邢颖慧, 程德, 张艳宁. 基于预训练的持续学习方法综述(特邀)[J]. 计算机工程, 2025, 51(10): 1-17.
[6]	钱榕, 李鑫, 刘晓豫, 张克君, 郭竞桧, 叶仲昆. 考虑禁言机制的谣言传播模型[J]. 计算机工程, 2024, 50(8): 372-378.
[7]	周春良, 刘仰光, 孟祥佩. 基于贝叶斯衍生分类器的社交网络用户影响力评价模型[J]. 计算机工程, 2024, 50(6): 394-400.
[8]	宫阿娟, 潘天荣. 多病种眼底疾病诊断的深度学习策略讨论[J]. 计算机工程, 2024, 50(5): 363-372.
[9]	李政学, 李枝名, 彭德中, 陈杰. 基于特征对比学习和图卷积的社交网络用户分类[J]. 计算机工程, 2024, 50(4): 258-266.
[10]	吴海鹏, 钱育蓉, 冷洪勇. 基于双向注意力机制的多模态关系抽取[J]. 计算机工程, 2024, 50(4): 160-167.
[11]	刘帅威, 李智, 王国美, 张丽. 基于Transformer和GAN的对抗样本生成算法[J]. 计算机工程, 2024, 50(2): 180-187.
[12]	靳雁霞, 史志儒, 杨晶, 刘亚变, 乔星宇, 张翎. 布料与精细建模物体间的碰撞检测算法研究[J]. 计算机工程, 2023, 49(7): 269-277.
[13]	陈锐, 孙羽菲, 郭强, 隋轶丞, 周振辉, 石昌青, 张玉志. OclDNN:一种可应用于TensorFlow的通用DNN库[J]. 计算机工程, 2023, 49(4): 138-148.
[14]	石磊, 张吉涛, 高宇飞, 卫琳, 陶永才. 基于Transformer与BiLSTM的网络流量入侵检测[J]. 计算机工程, 2023, 49(3): 29-36,57.
[15]	王春东, 孙嘉琪, 杨文军. 基于矫正理解的中文文本对抗样本生成方法[J]. 计算机工程, 2023, 49(2): 37-45.

选择文件类型/文献管理软件名称

选择包含的内容