A rights-based re-integration method for JPEG steganalysis

SUN Shou-jian; WEI Li-xian; LIU Jia; SU Guang-wei

doi:10.3788/YJYXS20153002.0326

您当前的位置：

首页 >

文章列表页 >

A rights-based re-integration method for JPEG steganalysis

Imaging Technology & Image Processing | 更新时间：2020-08-12

- A rights-based re-integration method for JPEG steganalysis
- Chinese Journal of Liquid Crystals and Displays Vol. 30, Issue 2, Pages: 326-332(2015)
- 作者机构：
  
  武警工程大学电子技术系网络与信息安全武警部队重点实验室,陕西西安,710086
- 作者简介：
- 基金信息：
- DOI：10.3788/YJYXS20153002.0326
  CLC： TP309
- Received：02 April 2014，
  
  Revised：22 May 2014，
  
  Published：05 April 2015
- 稿件说明：
移动端阅览
孙寿健, 魏立线, 刘佳, 苏光伟. 一种基于权重融合的JPEG隐写分析方法[J]. 液晶与显示, 2015,30(2): 326-332

SUN Shou-jian, WEI Li-xian, LIU Jia, SU Guang-wei. A rights-based re-integration method for JPEG steganalysis[J]. , 2015,30(2): 326-332
孙寿健, 魏立线, 刘佳, 苏光伟. 一种基于权重融合的JPEG隐写分析方法[J]. 液晶与显示, 2015,30(2): 326-332 DOI： 10.3788/YJYXS20153002.0326.

SUN Shou-jian, WEI Li-xian, LIU Jia, SU Guang-wei. A rights-based re-integration method for JPEG steganalysis[J]. , 2015,30(2): 326-332 DOI： 10.3788/YJYXS20153002.0326.

摘要

针对JPEG图像通用隐写检测中检测效率低、训练时间长的问题

提出一种基于集成分类器的新检测方法。算法以CC-PEV为特征对图像进行描述并作为隐写分析特征;然后

随机构造若干个特征子空间

用bootstrap方法构造图像训练子集

分别进行训练得到数个基分类器;根据基分类器的分类结果赋予基分类器不同的权重

将基分类器的结果按照其权重进行融合得到最终的结果。本文对该算法进行了测试

对它的集成性、检测准确率和训练时间进行分析。实验结果表明

相对于传统的集成方法

本文方法用自举方法构造训练集、随机方法构造子特征空间、赋予基分类器不同权重进行融合能够显著地提高算法准确率。本文方法相对于SVM和传统的集成分类方法

具有更高的检测率

对于特征维数更大的图像检测

具有更好的拓展性和一般适用性。

Abstract

Aiming at the problem of the high dimension features and complicated training in image universal steganalysis technology

this paper proposes a steganalysis method based on ensemble classifier. The method designs a ensemble classifier which combines the difference of base classifier and weighted fusion. Specifically

the algorithm uses CC-PEV as steganalysis features. Then

the method randomly structures several feature subspace

uses the bootstrap method to construct image training subsets

and train them to get a number of base classifiers. Finally

every base classifier is given weight that different from each other according to the classification result and then the final forecast is got after the combination of classification results with their respective weight. The experimental results show that using ensemble classifier for training and giving base classifiers with different weights fusion could effectively improve the accuracy of the algorithm.

关键词

Keywords

references

张军, 熊枫, 张丹. 图像隐写分析技术综述 [J]. 计算机工程, 2013, 39(4): 165-168, 172. Zhang J, Xiong F, Zhang D. Overview on image steganalysis technology [J]. Computer Engineering, 2013, 39(4): 165-168, 172. (in Chinese)[2] 王朔中, 张新鹏, 张卫明. 以数字图像为载体的隐写分析研究进展 [J]. 计算机学报, 2009, 32(7): 1247-1263. Wang S Z, Zhang X P, Zhang W M. Recent advances in image-based steganalysis research [J]. Chinese Journal of Computers, 2009, 32(7): 1247-1263. (in Chinese)[3] 韩晓东. JPEG 图像隐写分析技术研究 [D]. 郑州: 解放军信息工程大学, 2009. Han X D. Research on steganlysis for JPEG images [D]. Zhengzhou: PLA Information Engineering University, 2009. (in Chinese)[4] Kodovsky J,Fridrich J, Holub V. Ensemble classifiers for steganalysis of digital media [J]. IEEE Transactions on Information Forensics and Security, 2012, 7(2): 432-444.[5] Kodovský J, Fridrich J. Calibration revisited [C]. Proceedings of the 11th ACM Workshop on Multimedia and Security, 2009: 63-74.[6] 尹华. 面向高位和不平衡数据分类的集成学习研究 [D]. 武汉:武汉大学, 2012. Yin H. Research of ensemble learning for high-dimensional and imbalanced data classification [D]. Wuhan:Wuhan University, 2012. (in Chinese)[7] Kodovsk ý J, Fridrich J.Steganalysis of JPEG images using rich models [C]. International Society for Optics and Photonics, 2012: 83030A-83030A-13.[8] Kodovský J, Fridrich J.Steganalysis in high dimensions: Fusing classifiers built on random subspaces [C]. International Society for Optics and Photonics, 2011: 78800L-78800L-13.[9] 李开达. 隐藏信息检测综合评估及融合技术研究 [D]. 郑州: 解放军信息工程大学, 2012. Li K D.Research on comprehensive evaluation and fusion of steganalysis techniques [D]. Zhengzhou: PLA Information Engineering University, 2012. (in Chinese)[10] Westfeld A.F5—a steganographic algorithm [M]. Springer Berlin Heidelberg, 2001: 289-302.[11] Ullerich C, Westfeld A.Weaknesses of MB2 [M]. Springer Berlin Heidelberg, 2008: 127-142.[12] Fridrich J, Goljan M, Hogea D. Attacking the outguess [C]. Proceedings of the ACM Workshop on Multimedia and Security, 2002.[13] 徐茂. 基于SVM的JPEG图像隐写分析研究[D]. 长沙: 湖南大学, 2012. Xu M.Steganalysis for JPEG Image Based on SVM [D].Changsha: Hunan University, 2012. (in Chinese)

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Image steganalysis algorithm based on feature ranking

A key point detection method of 3 D face based on multi-feature

Image steganalysis based on bayesian classifier

Related Author

Xing-chun ZHANG

Shou-jian SUN

Chao FENG

Qing-jiang CHEN

Related Institution

Command of Heilongjiang Crops of PAP

Command of Jiamusi Detachment, Heilongjiang Crops of PAP

Network and Information Security Key Laboratory of PAP, Engineering University of the PAP

School of Resources & Surveying Engineering, Shaanxi Energy Institute

School of Science, Xi'an University of Architecture and Technology

AI问答

Address：No.3888 Dong Nanhu Road, Changchun, Jilin, China 130033 Postal code：130033
Tel：0431-86176059 Email：yjxs@ciomp.ac.cn
Technical support is provided by Beijing Founder electronics co., LTD 吉ICP备11002662号-17 京公网安备11010802024621
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