降低特征类内离散度的JPEG图像隐写分析

doi:10.3969/j.issn.0255-8297.2019.01.005

应用科学学报 ›› 2019, Vol. 37 ›› Issue (1): 41-50.doi: 10.3969/j.issn.0255-8297.2019.01.005

降低特征类内离散度的JPEG图像隐写分析

汪然^1,2, 牛少彰², 平西建¹, 张涛¹, 桑晓丹³

1. 信息工程大学信息系统工程学院, 郑州 450001;
2. 北京邮电大学计算机学院, 北京 100876;
3. 解放军31401部队, 济南 250002

收稿日期:2017-09-24 修回日期:2018-04-10 出版日期:2019-01-31 发布日期:2019-01-31
作者简介:汪然,博士,讲师,研究方向:图像处理、多媒体信息安全,E-mail:nemo2007@163.com;牛少彰,教授,研究方向:信息安全,E-mail:szniu@bupt.edu.cn;平西建,教授,博导,研究方向:图像处理、信息安全,E-mail:pingxijian@163.com
基金资助:
国家自然科学基金（No.61602511，No.61572518，No.U1636202）资助

Steganalysis of JPEG Images Based on Reducing Between-Class Scatter

WANG Ran^1,2, NIU Shao-zhang², PING Xi-jian¹, ZHANG Tao¹, SANG Xiao-dan³

1. Institute of Information System and Engineering, Information Engineering University, Zhengzhou 450001, China;
2. School of Computer Science & Technology, Beijing University of Posts and Telecommunications, Beijing 100876, China;
3. Troops 31401 PLA, Jinan 250002, China

Received:2017-09-24 Revised:2018-04-10 Online:2019-01-31 Published:2019-01-31

摘要/Abstract

摘要：

图像内容特征差异使得载体、载密图像的隐写检测特征混淆在一起而难以区分，这导致图像隐写分析成了一个"类内分散、类间聚合"的分类问题.针对此问题，从降低因图像内容、处理手段等造成的隐写检测特征类内离散度的角度出发，提出了一种更加可靠的隐写检测模型.依据内容复杂度将待检测图像分类，分别提取具有相同内容复杂度的每一类图像的隐写检测特征和训练分类器，得到最终检测结果.数据分析和实验结果表明：基于图像分类的隐写分析方法能够有效提高检测性能.

关键词: 图像分类, 隐写分析, 类内离散度, 图像内容复杂度

Abstract:

Compared with the process of embedding, image contents make a more significant impact on the differences of image statistical characteristics. This makes the image steganalysis to be a classification problem with bigger within-class scatter distances and smaller between-class scatter distances. In this paper, a new steganalysis framework which can reduce the differences of image statistical characteristics caused by various content and processing methods is proposed. The given images are classified according to the texture complexity. Steganalysis features are separately extracted from each subset with the same or close complexity evaluation function to build a classifier. The theoretical analysis and experimental results can demonstrate the validity of the proposed framework.

Key words: steganalysis, between-class scatter, image classification, image content complexity

中图分类号:

TP391

汪然, 牛少彰, 平西建, 张涛, 桑晓丹. 降低特征类内离散度的JPEG图像隐写分析[J]. 应用科学学报, 2019, 37(1): 41-50.

WANG Ran, NIU Shao-zhang, PING Xi-jian, ZHANG Tao, SANG Xiao-dan. Steganalysis of JPEG Images Based on Reducing Between-Class Scatter[J]. Journal of Applied Sciences, 2019, 37(1): 41-50.

参考文献

[1] Pevný T, Fridrich J. Merging Markov and DCT features for multi-class JPEG steganalysis[C]//Proceedings of SPIE, 2007, 6505:650503.
[2] Kodovský J, Feideich J. Calibration revisited[C]//Proceedings of ACM Workshop Multimedia and Security. Princeton, New Jersey, 2009:63-74.
[3] Pevný T, Bas P, Fridrich J. Steganalysis by subtractive pixel adjacency matrix[J]. IEEE Transaction on Information Forensics and Security, 2010, 5(2):215-224.
[4] Liu Q. Steganalysis of DCT-embedding based adaptive steganography and YASS[C]//Proceedings of 13th ACM Multimedia & Security Workshop. New York, ACM, 2011:77-86.
[5] Goljan M, Fridrich J, Holotyak T. New blind steganalysis and its implications[C]//Proceedings of SPIE, Electronic Imaging, Security, Steganography, and Watermarking of Multimedia. San Jose, CA, 2006:1-13.
[6] Wang Y, Moulin P. Optimized feature extraction for learning based image steganalysis[J]. IEEE Transaction on Information Forensics and Security, 2007, 2(1):31-45.
[7] Kodovský J, Fridrich J, Holub V. Ensemble classifiers for steganalysis of digital media[J]. IEEE Transaction on Information Forensics and Security, 2012, 7(2):432-444.
[8] Kodovský J, Fridrich J. Steganalysis of JPEG images using rich models[C]//Proceedings of SPIE, 2012, 8303:83030A.
[9] Holub V, Fridrich J, Denemark T. Random projections of residuals as an alternative to cooccurrences in steganalysis[C]//Proceedings of SPIE, Electronic Imaging, Media Watermarking, Security, and Forensics. San Francisco, CA, 2013:3-7.
[10] Cho S, Cha B, Gawecki M, Kuo C. Block-based image steganalysis:algorithm and performance evaluation[J]. Journal of Visual Communication and Image Representation, 2013, 24(7):846-856.
[11] Wang R, Xu M, Ping X, Zhang T. Steganalysis of JPEG images by block texture based segmentation[J]. Multimedia Tools Application, 2015, 74(15):5725-5746.
[12] 汪然,许漫坤,平西建,张涛. 基于分割的空域图像隐写分析[J]. 自动化学报,2014, 40(12):2936-2943. Wang R, Xu M K, Ping X J, Zhang T. Steganalysis of spatial images based on segmentation[J]. Acta Automatica Sinica, 2014, 40(12):2936-2943. (in Chinese)
[13] Fisher R A. The use of multiple measurements in taxonomic problems[J]. Annals of Eugenics, 1936, 7(2):179-188.
[14] Filler T, Pevný T, Bas P. BOSS[EB/OL].[2007-07-01]. http://boss.gipsa-lab.grenobleinp.fr/BOSSRank/.
[15] Bas P, Furon T. Bows-2[EB/OL].[2007-07-01]. http://bows2.gipsa-lab.inpg.fr/BOWS2OrigEp3.tgz.
[16] Goljan M, Fridrich J, Holotyak T. The USDA NRCS Photo Gallery[EB/OL].[2008-09-14]. http://photogallery.nrcs.usda.gov.
[17] Schaefer G, Stich M. UCID-an uncompressed colour image database[R]. UK:Nottingham Trent University, 2003.
[18] Fridrich J, Pevný T, Kodovský J. Statistically undetectable JPEG steganography:dead ends, challenges, and opportunities[C]//Proceedings of ACM Multimedia and Security Workshop. Texas, 2007:3-14.
[19] Westfeld A. High capacity despite better steganalysis (F5-a steganographic algorithm)[C]//Proceedings of 4th International Workshop on Information Hiding. Pittsburgh, PA, 2001:289-302.
[20] Sallee P. Model-based steganography[C]//Proceedings of International Workshop on Digital Watermarking. Seoul, Korea, 2003:154-167.
[21] Kim Y, Duric Z, Richards D. Modified matrix encoding technique for minimal distortion steganography[C]//Proceedings of International Workshop on Information Hiding. Heidelberg, 2007:314-327.

降低特征类内离散度的JPEG图像隐写分析

Steganalysis of JPEG Images Based on Reducing Between-Class Scatter

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	司广文, 秦川, 姚恒, 韩彦芳, 张志超. 基于纹理特征分类与合成的鲁棒无载体信息隐藏[J]. 应用科学学报, 2020, 38(3): 441-454.
[2]	李松斌, 杨洁, 刘鹏, 王凌睿. 基于关联网络模型的H.264/AVC运动矢量调制信息隐藏检测方法[J]. 应用科学学报, 2019, 37(5): 663-672.
[3]	卢才武, 齐凡, 阮顺领. 基于深度图像分析的细粒度矿石分级测定方法[J]. 应用科学学报, 2019, 37(4): 490-500.
[4]	宋海峰, 陈广胜, 景维鹏, 杨巍巍. 基于(2D)²PCA的受限玻尔兹曼机图像分类算法及其并行化实现[J]. 应用科学学报, 2018, 36(3): 495-503.
[5]	史晓裕, 李斌, 谭舜泉. 深度学习空域隐写分析的预处理层[J]. 应用科学学报, 2018, 36(2): 309-320.
[6]	方志坚, 傅仰耿, 陈建华. 纹理图像分类的置信规则库推理方法[J]. 应用科学学报, 2017, 35(5): 545-558.
[7]	郑国华, 冯国瑞, 余江, 程航, 张新鹏. 基于LSB检测的JPEG隐写分析特征增强方法[J]. 应用科学学报, 2016, 34(6): 670-676.
[8]	孙物一, 冯国瑞. 基于高维特征正则化的JPEG图像隐写分析[J]. 应用科学学报, 2016, 34(5): 555-563.
[9]	刘格, 黄方军, 李中华. 针对自适应隐写的通用隐写分析研究[J]. 应用科学学报, 2016, 34(5): 598-604.
[10]	欧阳春娟, 刘昌鑫, 刘欢. 混合蛙跳优化决策面的LSB±k隐写算法[J]. 应用科学学报, 2015, 33(6): 663-670.
[11]	谢凯，张涛，奚玲，李文祥，平西建. k均值聚类的混合异构图像隐写分析[J]. 应用科学学报, 2014, 32(5): 543-550.
[12]	闫晓蒙，张涛，李文祥，平西建. 结合WS残差法和MAP载体估计的隐写负载定位[J]. 应用科学学报, 2014, 32(4): 401-408.
[13]	李韬，平西建. 一种基于图像内容的隐写分析方法[J]. 应用科学学报, 2013, 31(4): 394-401.
[14]	李星1,2，张涛1，何赞园2，李文祥1，李开达1，平西建1. 结合重压缩检测的JPEG 图像多类隐写分析[J]. 应用科学学报, 2013, 31(2): 190-196.
[15]	李侃1,2，平西建1. 基于图像内容和特征融合的隐写盲检测[J]. 应用科学学报, 2013, 31(1): 97-103.