基于风格迁移纹理合成与识别的构造式信息隐藏
作者:
作者单位:

作者简介:

通讯作者:

秦川,E-mail:qin@usst.edu.cn

中图分类号:

TP391

基金项目:

国家自然科学基金(62172280,U20B2051,62172281);上海市科委高校能力建设项目(20060502300)


Constructive Data Hiding Based on Texture Synthesis and Recognition with Image Style Transfer
Author:
Affiliation:

Fund Project:

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

    传统的信息隐藏算法大都通过修改载体达到隐藏秘密信息的目的,但不可避免地会在载体数据中留下修改痕迹,故常难以抵抗隐写分析技术的检测,为此无载体信息隐藏应运而生。无载体信息隐藏并非不使用载体,而是不对载体数据进行修改。为了提高无载体信息隐藏算法的隐藏容量和鲁棒性,本文提出了一种基于风格迁移纹理合成与识别的构造式信息隐藏算法。该算法首先选取不同类别的自然图像和纹理图像分别建立内容图像库和纹理风格图像库,并根据内容图像库中自然图像的类别构建二进制码的映射字典;其次为了接收方能够从含密图像中提取出秘密信息,需要构建带标签的纹理图像库,并将其作为训练集输入·到卷积神经网络中,通过迭代训练获得纹理图像识别模型。在秘密信息隐藏时,根据秘密信息片段选择对应类别的自然图像,并按照一定的顺序组合成含密拼接图像,随后从纹理图像库中随机选择一张纹理图像,通过风格迁移的方法将含密拼接图像转换成含密纹理图像,从而完成秘密信息隐藏过程。在信息提取过程中,通过纹理图像识别模型可准确识别出含密纹理图像原本对应的图像类别,再对照映射字典即可提取出秘密信息。实验结果表明,本文算法生成的含密纹理图像具有良好的视觉效果,秘密信息隐藏容量较高,且对JPEG压缩、高斯噪声等攻击具有较强的鲁棒性。

    Abstract:

    Most traditional information hiding methods embed secret data by modifying cover data,which will leave traces of modification on cover data and will be difficult to resist the detection of the existing steganalysis algorithms.Consequently,the technique of coverless information hiding emerges,which hide secret data without modifying cover data.In order to improve the hiding capacity and robustness of coverless information hiding,in this paper,a constructive data hiding method based on texture synthesis and recognition with image style transfer is proposed.Firstly,a number of natural images and texture images with different categories are used to construct the content image database and the textural style image database,respectively.A mapping dictionary of binary codes is established according to the categories of natural images in the content image database.Secondly,in order to ensure data extraction correctly,the labeled textural image database should be constructed and inputted into the convolutional neural network as a training database,thus,the texture image recognition model can be obtained by iterative training.During secret data hiding,natural images are selected from the content image database according to to-be-embedded secret data fragments,which compose of a stego mosaic image.Then,a texture image is selected from the textural style image database,and the stego texture image can be generated based on the selected texture image and the stego mosaic image with the strategy of style transfer.During secret data extraction,our texture image recognition model can accurately identify the original categories of stego texture images corresponding to natural images,and secret data can be finally obtained by the mapping dictionary.Experimental results demonstrate that our method achieves satisfactory visual effect of stego texture image,high hiding capacity and strong robustness,which is superior to some state-of-the-art methods.

    参考文献
    相似文献
    引证文献
引用本文

秦川,董腾林,姚恒.基于风格迁移纹理合成与识别的构造式信息隐藏.软件学报,,():0

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2021-05-31
  • 最后修改日期:2021-10-08
  • 录用日期:
  • 在线发布日期: 2022-10-26
  • 出版日期:
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