基于生成对抗网络的空域彩色图像隐写失真函数设计方法
作者:
作者单位:

作者简介:

廖鑫,E-mail:xinliao@hnu.edu.cn

通讯作者:

廖鑫,E-mail:xinliao@hnu.edu.cn

基金项目:

国家自然科学基金项目(61972142,61872356);国家重点研发计划课题(2019QY(Y)0207,2019QY2202);湖南省自然科学基金项目(2020JJ4212);信息网络安全公安部重点实验室开放课题(C20611).


Steganographic Distortion Function Learning Method for Spatial Color Image Based on Generative Adversarial Network
Author:
Affiliation:

Fund Project:

This work is supported by National Natural Science Foundation of China (61972142, 61872356), National Key Research and Development Program of China (2019QY(Y)0207, 2019QY2202), Hunan Provincial Natural Science Foundation of China (2020JJ4212), the Key Lab of Information Network Security and the Ministry of Public Security of China (C20611).

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
    摘要:

    自适应隐写是图像隐写方向的研究热点,它通过有效地设计隐写失真函数,自适应地将秘密信息隐藏在图像复杂的纹理区域,具有很强的隐蔽性.近年来,基于生成对抗网络的隐写失真函数设计研究在空域灰度图像上已经取得了突破性的进展,但是目前还没有针对空域彩色图像的研究.与灰度图像相比,彩色图像隐写需要考虑保护RGB通道间相关性,同时合理地分配RGB三个通道的嵌密容量.本文设计了一个基于生成对抗网络设计空域彩色图像隐写失真函数的框架CIS-GAN(color image steganography based on generative adversarial network),生成器网络采用两个U-Net子网络结构,第一个U-Net子网络生成修改概率矩阵,第二个U-Net子网络进行正负向修改概率调节,有效地降低对彩色图像通道相关性的破坏.针对彩色图像载体,修改灰度图像隐写分析器作为网络的对抗部分.在生成器损失函数中对彩色图像三个通道总的隐写容量进行控制,生成器能够自动学习分配三个通道嵌密容量.实验结果表明,与现有彩色图像隐写失真函数设计方法相比,本文提出的网络结构能够更好地抵抗彩色图像隐写分析器的检测.

    Abstract:

    Adaptive image steganography has been becoming a hot topic, as it conceals covert information within the texture region of an image by employing a defined distortion function, which guarantees remarkable security. In spatial gray-scale image steganography, the research on automatically generating steganographic distortion using the generative adversarial network has achieved a significant breakthrough recently. However, to the best of our knowledge, there are not related works in spatial color image steganography. Compared with the gray-scale image, color image steganography should preserve the channel correlation and reasonably assign the embedding capacity among RGB channels simultaneously. This paper first proposes a framework based on generative adversarial network to automatically learn to generate the steganographic distortion for spatial color image, which is termed as CIS-GAN (color image steganography based on generative adversarial network). The generator is composed of two U-Net subnetworks, one of two subnetworks translates a cover image into a modification probability map which is the sum of positive/negative modification probability, while another one learns the proportion of positive modification probability. The structure of the designed generator can effectively preserve RGB channels correlation, so as to enhance the steganography security. Also, the generator can automatically learn to allocate the embedding capacity for three channels via controlling the total steganographic capacity in generator's loss function and alternately training the discriminator. The experimental results show that our proposed framework outperforms the advanced spatial color image steganographic schemes in resisting the color image steganalysis.

    参考文献
    相似文献
    引证文献
引用本文

廖鑫,唐志强,曹纭.基于生成对抗网络的空域彩色图像隐写失真函数设计方法.软件学报,,():0

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
历史
  • 收稿日期:2020-08-11
  • 最后修改日期:2020-10-12
  • 录用日期:
  • 在线发布日期: 2021-04-21
您是第位访问者
版权所有:中国科学院软件研究所 京ICP备05046678号-3
地址:北京市海淀区中关村南四街4号,邮政编码:100190
电话:010-62562563 传真:010-62562533 Email:jos@iscas.ac.cn
技术支持:北京勤云科技发展有限公司

京公网安备 11040202500063号