Robust generative image steganography method based on distribution-preserving embedding and orthogonal mapping

Yuan Chengsheng; Wang Haoyu; Yin Qingwei; Cao Yi; Liu Qingcheng; Fu Zhangjie

doi:10.11959/j.issn.1000-436x.2026063

Chinese

您当前的位置：

首页 >

文章列表页 >

Robust generative image steganography method based on distribution-preserving embedding and orthogonal mapping

Correspondences | 更新时间：2026-05-07

- Robust generative image steganography method based on distribution-preserving embedding and orthogonal mapping
- Journal on Communications Vol. 47, Issue 4, Pages: 288-298(2026)
- 作者机构：
  
  1.南京信息工程大学计算机学院、网络空间安全学院，江苏南京 210044
  2.南京信息工程大学数字取证教育部工程研究中心，江苏南京 210044
  3.江苏省教育厅教育宣传中心，江苏南京 210036
  4.无锡学院物联网工程学院，江苏无锡 214105
- 作者简介：
- 基金信息：
  
  The National Natural Science Foundation of China(U22B2062;U23B2023;62372125)
- DOI：10.11959/j.issn.1000-436x.2026063
  CLC： TP391
- Received：30 December 2025，
  
  Revised：2026-03-01，
  
  Accepted：02 March 2026，
  
  Published：20 April 2026
- 稿件说明：
移动端阅览
袁程胜,王昊宇,尹青伟等.基于分布保持嵌入与正交映射的鲁棒生成式图像隐写方法[J].通信学报,2026,47(04):288-298.

Yuan Chengsheng,Wang Haoyu,Yin Qingwei,et al.Robust generative image steganography method based on distribution-preserving embedding and orthogonal mapping[J].Journal on Communications,2026,47(04):288-298.
袁程胜,王昊宇,尹青伟等.基于分布保持嵌入与正交映射的鲁棒生成式图像隐写方法[J].通信学报,2026,47(04):288-298. DOI： 10.11959/j.issn.1000-436x.2026063.

Yuan Chengsheng,Wang Haoyu,Yin Qingwei,et al.Robust generative image steganography method based on distribution-preserving embedding and orthogonal mapping[J].Journal on Communications,2026,47(04):288-298. DOI： 10.11959/j.issn.1000-436x.2026063.

摘要

在协同多智能体系统的隐蔽通信中，基于生成式模型的隐写技术能够直接在公开信道中合成载体图像，为安全传输指令或数据提供了新途径。针对现有生成式隐写方法无法兼顾鲁棒性、载体图像质量与抗检测性等问题，提出了一种基于分布保持嵌入与正交映射的生成式图像隐写方法。首先，设计一种分布保持的信息嵌入机制，将秘密信息编码至采样的高斯潜向量中。其次，引入向量重构模块，生成满足标准正态分布的向量作为模型输入，以增强对信道干扰的鲁棒性及抗隐写分析能力。最后，接收方基于共享的密钥执行逆向操作以提取信息。实验结果表明，在不同隐写容量下，所提方法生成的载体图像均具有较好的视觉保真度，在遭受JPEG压缩、高斯噪声等常见信道攻击后，信息提取准确率仍保持在97%以上，表现出较强的鲁棒性。此外，面对常见的隐写分析检测器，误检率均保持在0.500左右，表现出优异的抗隐写分析性能。

Abstract

In covert communication for collaborative multi-agent systems

generative model-based image steganography can directly synthesize stego-images over public channels

offering a novel approach for secure transmission of instructions and data. Given that existing generative steganography methods can’t simultaneously achieve robustness

cover image quality

and resistance to detection

a generative image steganography method based on distribution-preserving embedding and orthogonal mapping was proposed. Firstly

a distribution-preserving information embedding mechanism was designed to embed the secret information into the sampled Gaussian latent vector. Secondly

a vector reconstruction module was introduced to generate inputs that follow the standard normal distribution

thereby enhancing robustness against channel interference and resistance to steganalysis. Finally

the receiver performed inverse operations based on the shared key to extract the information. Experimental results show that

under different steganographic capacities

the cover images generated by the proposed method consistently achieve superior visual fidelity. After attacks such as JPEG compression and Gaussian noise

the information extraction accuracy remains above 97%

demonstrating strong robustness. Against common steganalysis detectors

the false positive rate is approximately 0.500

highlighting excellent detectability resistance.

关键词

Keywords

references

Zhou X J , Peng W L , Yang B Y , et al . Linguistic steganography based on adaptive probability distribution [J ] . IEEE Transactions on Dependable and Secure Computing , 2022 , 19 ( 5 ): 2982 - 2997 .

Deng Y X , Tan Y , Mao L , et al . A robust generative coverless co-steganography method of text and image [J ] . International Journal of Autonomous and Adaptive Communications Systems , 2025 , 18 ( 6 ): 485 - 508 .

Gopalan K . Audio steganography using bit modification [C ] // Procee-dings of the 2003 International Conference on Multimedia and Expo . Piscataway : IEEE Press , 2003 : I-629.

周志立 , 丁淳 , 李进 , 等 . 生成式隐写研究 [J ] . 计算机学报 , 2023 , 46 ( 9 ): 1855 - 1887 .

Zhou Z L , Ding C , Li J , et al . Research on generative steganography [J ] . Chinese Journal of Computers , 2023 , 46 ( 9 ): 1855 - 1887 .

Duan X T , Song H X , Qin C , et al . Coverless steganography for digital images based on a generative model [J ] . Computers , Materials & Continua, 2018 , 55 ( 3 ): 483 - 493 .

Lee W K , Ong S , Wong K , et al . A novel coverless information hiding technique using pattern image synthesis [C ] // Proceedings of the 2018 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) . Piscataway : IEEE Press , 2018 : 1122 - 1127 .

Cao Y , Zhou Z L , Wu Q M J , et al . Coverless information hiding based on the generation of anime characters [J ] . EURASIP Journal on Image and Video Processing , 2020 , 2020 : 36 .

Wei P , Li S , Zhang X P , et al . Generative steganography network [C ] // Proceedings of the 30th ACM International Conference on Multimedia . New York : ACM , 2022 : 1621 - 1629 .

Kim D , Shin C , Choi J , et al . Diffusion-Stego: Training-free diffusion generative steganography via message projection [J ] . Information Sciences , 2025 , 718 : 122358 .

Yang Z J , Chen K J , Zeng K , et al . Provably secure robust image steganography [J ] . IEEE Transactions on Multimedia , 2024 , 26 : 5040 - 5053 .

吴槟 , 薛瑞 . 基于深度学习特征分布优化的无载体图像隐写方法 [J ] . 信息安全学报 , 2024 : doi.10.19363/J.cnki.cn10-1380/tn.2024.04.05.

Wu B , Xue R . A coverless image steganography method using deeplearning with feature distribution optimization [J ] . Journal of Cyber Security , 2024 : doi.10.19363/J.cnki.cn10-1380/tn.2024.04.05.

Mielikainen J . LSB matching revisited [J ] . IEEE Signal Processing Letters , 2006 , 13 ( 5 ): 285 - 287 .

Filler T , Judas J , Fridrich J . Minimizing additive distortion in steganography using syndrome-trellis codes [J ] . IEEE Transactions on Information Forensics and Security , 2011 , 6 ( 3 ): 920 - 935 .

Li W X , Zhang W M , Li L , et al . Designing near-optimal steganographic codes in practice based on polar codes [J ] . IEEE Transactions on Communications , 2020 , 68 ( 7 ): 3948 - 3962 .

Sharifzadeh M , Aloraini M , Schonfeld D . Adaptive batch size image merging steganography and quantized Gaussian image steganography [J ] . IEEE Transactions on Information Forensics and Security , 2020 , 15 : 867 - 879 .

Sharp T . An implementation of key-based digital signal steganography [C ] // Information Hiding . Berlin , Springer , 2001 : 13 - 26 .

Wani M A , Sultan B . Deep learning based image steganography: a review [J ] . WIREs Data Mining and Knowledge Discovery , 2023 , 13 ( 3 ): e1481 .

Zhu J R , Kaplan R , Johnson J , et al . HiDDeN: hiding data with deep networks [C ] // Computer Vision - ECCV 2018 . Cham : Springer International Publishing , 2018 : 682 - 697 .

Peng Y Y , Hu D H , Wang Y F , et al . StegaDDPM: generative image steganography based on denoising diffusion probabilistic model [C ] // Proceedings of the 31st ACM International Conference on Multimedia . New York : ACM , 2023 : 7143 - 7151 .

Zhou Z L , Sun H Y , Harit R , et al . Coverless image steganography without embedding [C ] // Cloud Computing and Security . Cham : Springer International Publishing , 2015 : 123 - 132 .

Holub V , Fridrich J , Denemark T . Universal distortion function for steganography in an arbitrary domain [J ] . EURASIP Journal on Information Security , 2014 , 2014 : 1 .

Yang X , Wu T Y , Huang F J . Reversible data hiding in JPEG images based on coefficient-first selection [J ] . Signal Processing , 2022 , 200 : 108639 .

Hu X X , Li S , Ying Q C , et al . Establishing robust generative image steganography via popular stable diffusion [J ] . IEEE Transactions on Information Forensics and Security , 2024 , 19 : 8094 - 8108 .

Cachin C . An information-theoretic model for steganography [J ] . Information and Computation , 2004 , 192 ( 1 ): 41 - 56 .

Zhou Z L , Su Y C , Li J , et al . Secret-to-image reversible transformation for generative steganography [J ] . IEEE Transactions on Dependable and Secure Computing , 2023 , 20 ( 5 ): 4118 - 4134 .

Xu Y H , Sun W , Tang C P , et al . Security-robustness trade-offs in diffusion steganography: a comparative analysis of pixel-space and vae-based architectures [PP ] . V2 . ( 2025-10-08 )[ 2025-12-30 ] . arXiv: arXiv. 2510 . 07219 .

Zhang X , Song T H , Peng F , et al . Denoising diffusion probabilistic steganography based on standardized secret noise [J ] . IEEE Signal Processing Letters , 2025 , 32 : 4124 - 4128 .

Zhou Z L , Dong X H , Meng R H , et al . Generative steganography via auto-generation of semantic object contours [J ] . IEEE Transactions on Information Forensics and Security , 2023 , 18 : 2751 - 2765 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Radio frequency fingerprint data augmentation for indoor localization based on diffusion model

Related Author

Fu Zhangjie

Yuan Chengsheng

Weike ZENG

Hanxiao CHANG

Jinying XU

Jingjie TAO

Haojun AI

Related Institution

Engineering Research Center of Digital Forensics Ministry of Education, NUIST

School of Cyber Science and Engineering, Wuhan University

Key Laboratory of Aerospace Information Security and Trusted Computing Ministry of Education, Wuhan University

AI问答

Postal code：100079
Tel：（010）53879206 Email：tmw@bjxintong.com.cn
Technical support is provided by Beijing Founder electronics co., LTD 京ICP备09082226号-64 京公网安备11010602201714号
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰