ISEM：基于生成多对抗网络的图像隐写增强模型

周赛星; 叶苗欣; 骆伟祺

doi:10.11959/j.issn.2096-109x.2025053

您当前的位置：

首页 >

文章列表页 >

ISEM：基于生成多对抗网络的图像隐写增强模型

学术论文 | 更新时间：2025-11-08

- ISEM：基于生成多对抗网络的图像隐写增强模型
- ISEM：image steganography enhancement model based on generative multi-adversarial network
- 网络与信息安全学报 2025年11卷第5期页码：126-136
- 作者机构：
  
  1.中山大学计算机学院，广东广州510006
  2.广东省信息安全技术重点实验室，广东广州510006
- 作者简介：
  
  [ "周赛星（2001- ），女，广东深圳人，中山大学硕士生，主要研究方向为图像隐写。" ]
  [ "叶苗欣（1999- ），女，广东珠海人，中山大学硕士生，主要研究方向为图像隐写。" ]
  [ "骆伟祺（1980- ），男，广东广州人，博士，中山大学教授、博士生导师，主要研究方向为数字多媒体取证、图像隐写和图像隐写分析。" ]
- 基金信息：
  
  国家自然科学基金(62472458);广东省自然科学基金(2025A1515012823)
- DOI：10.11959/j.issn.2096-109x.2025053
  中图分类号： TP393
- 收稿：2024-12-31，
  
  修回：2025-03-19，
  
  纸质出版：2025-10-25
- 稿件说明：
移动端阅览
周赛星, 叶苗欣, 骆伟祺. ISEM:基于生成多对抗网络的图像隐写增强模型[J]. 网络与信息安全学报, 2025, 11(5): 126-136.

ZHOU S X, YE M X, LUO W Q. ISEM:image steganography enhancement model based on generative multi-adversarial network[J]. Chinese Journal of Network and Information Security, 2025, 11(5): 126-136.
周赛星, 叶苗欣, 骆伟祺. ISEM:基于生成多对抗网络的图像隐写增强模型[J]. 网络与信息安全学报, 2025, 11(5): 126-136. DOI： 10.11959/j.issn.2096-109x.2025053.

ZHOU S X, YE M X, LUO W Q. ISEM:image steganography enhancement model based on generative multi-adversarial network[J]. Chinese Journal of Network and Information Security, 2025, 11(5): 126-136. DOI： 10.11959/j.issn.2096-109x.2025053.

摘要

图像隐写技术根据在同一嵌入单元内

https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=93164161&type=

2.20133328

https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=93164178&type=

3.55599999

和

https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=93164152&type=

2.28600001

https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=93164165&type=

3.47133350

的嵌入代价是否相等，通常分为对称隐写和非对称隐写。对称隐写技术未能充分利用相邻像素之间的相互关系，限制了隐写安全性的提升。非对称隐写技术虽能显著增强隐写的安全性，但在生成非对称嵌入代价的过程中，现有方法往往存在训练过程不稳定、依赖人工经验或预设参数等问题，难以确保隐写的安全性。针对上述问题，提出了一种基于生成对抗网络的图像隐写增强模型（image steganography enhancement model

ISEM）。生成器网络采用相对轻量级的双分支网络结构，判别器网络则由3个基于深度学习的隐写分析网络组成。ISEM的训练分为两个阶段。第一阶段学习载体图像与现有嵌入代价之间的映射关系；第二阶段通过对抗训练调整初步学习的嵌入代价，生成高安全性的非对称嵌入代价。此外，在生成器中引入几何均值损失，在判别器中引入均值绝对偏差损失，以进一步优化训练效果。实验结果表明，ISEM能够有效增强对称（以及非对称）隐写方法的安全性，其相较于现有的非对称隐写方法，具有更佳的平均安全性。

Abstract

Image steganography techniques were typically categorized into symmetric and asymmetric types based on whether the embedding costs of +1 and -1 within the same embedding unit were equal. The interrelationship between adjacent pixels was not fully utilized by symmetric steganography techniques

which limited the enhancement of steganographic security. Although asymmetric steganography techniques were able to significantly improve security

existing methods were often faced with issues such as unstable training processes

reliance on manual experience

or dependence on preset parameters when generating asymmetric embedding costs

making it difficult to ensure steganographic security. To address these issues

an image steganography enhancement model (ISEM) based on a generative multi-adversarial network was proposed. A relatively lightweight dual-branch structure was adopted by the generator network

while the discriminator network was composed of three deep learning-based steganalysis networks. The training of ISEM was divided into two stages. In the first stage

the mapping relationship between the carrier image and existing embedding costs was learned. In the second stage

the initially learned embedding costs were adjusted through adversarial training to generate high-security asymmetric embedding costs. Additionally

a geometric mean loss was introduced in the generator

and a mean absolute deviation loss was incorporated in the discriminator to further optimize the training effect. Experimental results show that the security of both symmetric and asymmetric steganography methods are effectively enhanced by ISEM. Compared to existing asymmetric steganography methods

the highest average security is achieved by the proposed approach.

关键词

Keywords

references

LI B , HE J H , HUANG J W , et al . A survey on image steganography and steganalysis [J ] . Journal of Information Hiding and Multimedia Signal Processing , 2011 , 2 ( 2 ): 142 - 172 .

竺乐庆 , 郭钰 , 莫凌强 , 等 . DGANS: 基于双重生成式对抗网络的稳健图像隐写模型 [J ] . 通信学报 , 2020 , 41 ( 1 ): 125 - 133 .

ZHU L Q , GUO Y , MO L Q , et al . DGANS: robustness image steganography model based on double GAN [J ] . Journal on Communications , 2020 , 41 ( 1 ): 125 - 133 .

翟黎明 , 嘉炬 , 任魏翔 , 等 . 深度学习在图像隐写术与隐写分析领域中的研究进展 [J ] . 信息安全学报 , 2018 , 3 ( 6 ): 2 - 12 .

ZHAI L M , JIA J , REN W X , et al . Recent advances in deep learning for image steganography and steganalysis [J ] . Journal of Cyber Security , 2018 , 3 ( 6 ): 2 - 12 .

FRIDRICH J , FILLER T . Practical methods for minimizing embedding impact in steganography [C ] // Proceedings of the Security, Steganography, and Watermarking of Multimedia Contents IX . 2007 : 650502 .

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 .

KER A D , PEVNÝ T , KODOVSKÝ J , et al . The square root law of steganographic capacity [C ] // Proceedings of the 10th ACM Workshop on Multimedia and Security . New York : ACM , 2008 : 107 - 116 .

PEVNÝ T , FILLER T , BAS P . Using high-dimensional image models to perform highly undetectable steganography [C ] // Information Hiding . Berlin : Springer , 2010 : 161 - 177 .

HOLUB V , FRIDRICH J . Designing steganographic distortion using directional filters [C ] // Proceedings of the 2012 IEEE International Workshop on Information Forensics and Security (WIFS) . Piscataway : IEEE Press , 2012 : 234 - 239 .

LI B , WANG M , HUANG J W , et al . A new cost function for spatial image steganography [C ] // Proceedings of the 2014 IEEE International Conference on Image Processing (ICIP) . Piscataway : IEEE Press , 2014 : 4206 - 4210 .

GOODFELLOW I , POUGET-ABADIE J , MIRZA M , et al . Generative adversarial networks [J ] . Communications of the ACM , 2020 , 63 ( 11 ): 139 - 144 .

TANG W X , TAN S Q , LI B , et al . Automatic steganographic distortion learning using a generative adversarial network [J ] . IEEE Signal Processing Letters , 2017 , 24 ( 10 ): 1547 - 1551 .

YANG J H , RUAN D Y , HUANG J W , et al . An embedding cost learning framework using GAN [J ] . IEEE Transactions on Information Forensics and Security , 2019 , 15 : 839 - 851 .

RONNEBERGER O , FISCHER P , BROX T . U-Net: convolutional networks for biomedical image segmentation [C ] /// Proceedings of Medical Image Computing and Computer-Assisted Intervention . Berlin : Springer , 2015 : 234 - 241 .

TANG W X , LI B , BARNI M , et al . An automatic cost learning framework for image steganography using deep reinforcement learning [J ] . IEEE Transactions on Information Forensics and Security , 2020 , 16 : 952 - 967 .

HUANG D X , LUO W Q , LIU M L , et al . Steganography embedding cost learning with generative multi-adversarial network [J ] . IEEE Transactions on Information Forensics and Security , 2023 , 19 : 15 - 29 .

LI B , WANG M , LI X L , et al . A strategy of clustering modification directions in spatial image steganography [J ] . IEEE Transactions on Information Forensics and Security , 2015 , 10 ( 9 ): 1905 - 1917 .

DENEMARK T , FRIDRICH J . Improving steganographic security by synchronizing the selection channel [C ] // Proceedings of the 3rd ACM Workshop on Information Hiding and Multimedia Security . New York : ACM , 2015 : 5 - 14 .

WANG Z C , LV J P , WEI Q D , et al . Distortion function for spatial image steganography based on the polarity of embedding change [C ] // Proceedings of Digital Forensics and Watermarking . Berlin : Springer , 2017 : 487 - 493 .

GOODFELLOW I J , SHLENS J , SZEGEDY C , et al . Explaining and harnessing adversarial examples [EB ] . 2014 .

TANG W X , LI B , TAN S Q , et al . CNN-based adversarial embedding for image steganography [J ] . IEEE Transactions on Information Forensics and Security , 2019 , 14 ( 8 ): 2074 - 2087 .

MO H X , SONG T T , CHEN B L , et al . Enhancing JPEG steganography using iterative adversarial examples [C ] // Proceedings of the 2019 IEEE International Workshop on Information Forensics and Security (WIFS) . Piscataway : IEEE Press , 2019 : 1 - 6 .

LIU M L , LUO W Q , ZHENG P J , et al . A new adversarial embedding method for enhancing image steganography [J ] . IEEE Transactions on Information Forensics and Security , 2021 , 16 : 4621 - 4634 .

HUANG D X , LUO W Q , ZHENG P J , et al . Automatic asymmetric embedding cost learning via generative adversarial networks [C ] // Proceedings of the 31st ACM International Conference on Multimedia . New York : ACM , 2023 : 8316 - 8326 .

BOROUMAND M , CHEN M , FRIDRICH J . Deep residual network for steganalysis of digital images [J ] . IEEE Transactions on Information Forensics and Security , 2019 , 14 ( 5 ): 1181 - 1193 .

DENG X Q , CHEN B L , LUO W Q , et al . Fast and effective global covariance pooling network for image steganalysis [C ] // Proceedings of the ACM Workshop on Information Hiding and Multimedia Security . New York : ACM , 2019 : 230 - 234 .

ZHANG R , ZHU F , LIU J Y , et al . Depth-wise separable convolutions and multi-level pooling for an efficient spatial CNN-based steganalysis [J ] . IEEE Transactions on Information Forensics and Security , 2019 , 15 : 1138 - 1150 .

BAS P , FILLER T , PEVNÝ T . "Break Our Steganographic System": the ins and outs of organizing BOSS [C ] // Information Hiding . Berlin : Springer , 2011 : 59 - 70 .

FRIDRICH J , KODOVSKY J . Rich models for steganalysis of digital images [J ] . IEEE Transactions on Information Forensics and Security , 2012 , 7 ( 3 ): 868 - 882 .

XU G S , WU H Z , SHI Y Q . Structural design of convolutional neural networks for steganalysis [J ] . IEEE Signal Processing Letters , 2016 , 23 ( 5 ): 708 - 712 .

YEDROUDJ M , COMBY F , CHAUMONT M . Yedroudj-Net: an efficient CNN for spatial steganalysis [C ] // Proceedings of the 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . Piscataway : IEEE Press , 2018 : 2092 - 2096 .

BAS P , FURON T . Image database of BOWS-2 [J ] . Accessed , 2017 , 20 : 2016 - 2017 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

融合小波快捷连接生成对抗网络的面部性别伪造

基于生成对抗网络的隐私增强联邦学习方案

图像非加性隐写综述

生成对抗网络研究综述

基于生成对抗网络的文本序列数据集脱敏