PDF下载 分享
[1]杨 雪,杨 榆,雷 敏.基于SPAM 和特征优化的通用隐写分析算法改进[J].成都信息工程大学学报,2016,(01):65-69.
　YANG Xue,YANG Yu,LEI Min.Improvement of Universal Steganalysis based on SPAM and Feature Optimization[J].Journal of Chengdu University of Information Technology,2016,(01):65-69.
点击复制

基于SPAM 和特征优化的通用隐写分析算法改进

成都信息工程大学学报[ISSN:1006-6977/CN:61-1281/TN] 卷: 期数: 2016年01期 页码: 65-69 栏目: 电子信息科学与技术 出版日期: 2016-01-20
Title:
Improvement of Universal Steganalysis based on SPAM and Feature Optimization
文章编号:
2096-1618(2016)01-0065-05
作者:
杨 雪1;  杨 榆2;  雷 敏2
(1. 北方工业大学电子信息工程学院,北京100144;2. 北京邮电大学信息安全中心,北京100876)
Author(s):
YANG Xue1;  YANG Yu2;  LEI Min2
(1. Electronic Information Engineering Institute, North China University of Technology, Beijing 100144,China;2. Information Security Center, BeijingUniversity of Posts and Telecommunications, Beijing 100876,China)
关键词:
通用隐写分析; SPAM; Fisher score; 维度规约
Keywords:
universal steganalysis;  SPAM;  fisher score;  statute of the dimension
分类号:
TP309. 2
文献标志码:
A
摘要:
通用隐写分析特征的高维化趋势加剧,导致算法时间复杂度和空间复杂度急速上升。因此,研究如何在 维持检测率水平的同时降低特征维度,对隐写分析的实用性有重要意义。本研究通过主成分分析确定特征矢量的 最优维度;借用费希尔线性判别式思想,以“类内聚合冶和“类间离散冶程度评价各维特征区分自然和隐写载体的能 力,进而选取最优子集。分析针对主流通用检测模型的基础———SPAM 模型进行,仿真实验证明,优化子集具有良 好检测性和较低计算复杂度。
Abstract:
The tendency for high-dimension of universal steganalysis characteristics toward intensifying, and lead to the rapid rise in complexity of algorithm in time and space domain. So maintain the level of detection rates, and reduce the dimension of features at the same time, have significance in research of steganalysis. This paper determine the optimal dimension of feature vectors by principal component analysis; using the concept of Fisher linear discriminant, with the degree of "aggregations within class" and " discreteness between classes" to evaluate the ability of each dimensionfea- turesto distinguish natural and hidden carrier, and then select the optimal subset. The analysis directs atthe mainstream universal steganalysis model———SPAM model, and the simulation results show that optimal subset has a good detection and low computational complexity.

参考文献/References:

[1] 黄炜,赵险峰,冯登国,等. 基于主成分分析进行 特征融合的JPEG 隐写分析[ J]. 软件学报, 2012, 23(7): 1869-1879.
[2] Fridrich J, Goljan M. Practical steganalysis of dig-ital images-state of the art [C]. Security and Wa- termarking of Multimedia Contents, USA: SPIE 2001: 1-13.
[3] Pevny忆T, Bas P, Fridrich J. Steganalysis by sub- tractive pixel adjacency matrix [ J]. IEEE Trans- actions on Information Forensic and Security, 2010, 5(2): 215-224.
[4] 侯瑜. 图像隐写分析[D]. 北京:北京交通大学, 2015.
[5] HongW, T S Chen. A novel data embedding meth- od using adaptive pixel pair matching [ J]. IEEE Transaction on Information Forensics and Security, 2012, 7(1):176-184.
[6] 汪然,许漫坤,平西建,等. 基于分割的空域图像 隐写分析[ J]. 自动化学报, 2014, 40 ( 12): 2936-2943.
[7] Kodovsky J, Pevny T, Fridrich J. Modern Stega- nalysis CanDetect YASS[ C]. Proc. Of Conf. on Electronic Imaging, Media Forensics and Security. San Jose, USA: SPIE, 2010.
[8] H Zhang,XJ Ping,XU Mankun,et al. Steganalysis by subtractive pixel adjacency matrix and dimen- sionality reduction [ J]. Sciece China Information Sciences, 2014, 57(4): 1-7.
[9] Veenu Bhasin, Punam Bedi, Anuradha Singhal: Feature selection for steganalysis based on modified stochastic diffusion search using fisher score[ J]. International Conference on Advances in Compu- ting, Communications and Informatics (ICACCI). 2014: 2323-2330.
[10] ZHANG Xue-Qin, GU Chun-Hua. A method to ex- tract network intrusion detection feature [J]. Jour- nal of South China University of Technology (Natu- ral Science Edition). 2010, 38(1): 81-86.
[11] R G van Schyndel, A Z Tirkel, C. F. Osborne. A digital watermark[ J]. The IEEE International Conference on Image Processing. Austin, Texas, USA. 1994,(2): 86-90.
[12] Sharp T. An Implementation of Key-based Digit- al Signal Steganography[C]. / / In Proceedings of the 4th International Workshop on Information Hiding, London, UK, 2001, 13-26.
[13] W Luo, F Huang, J Huang. Edge adaptive image steganography based on LSB matching revisited [J]. IEEE Trans. Inform. Forensics Security, 2010, 5(2): 201 – 214.
[14] T Filler, T Pevny, P Bas. BOSS ( Break Our Steganography System) [EB/ OL]. Available:ht- tp: / / agents. fel. cvut. cz/ stegodata/ BossBase-1. 01-cover. tar. bz2, 2014.
[15] J Kodovsky, J Fridrich, V Holub. Ensemble classifiers forsteganalysis of digital media [ J]. IEEE Transactions on Information Forensicsand Security, 2012, 7(2): 432-444.

相似文献/References:

[1]段怀锋,杨 榆,雷 敏,等.基于训练集构造的图像通用盲检测算法改进[J].成都信息工程大学学报,2016,(01):70.
　DUAN Huai-feng,YANG Yu,LEI Min,et al.Improvement of Image Universal Blind Detection based on Training Set Construction[J].Journal of Chengdu University of Information Technology,2016,(01):70.

备注/Memo

收稿日期:2016-02-02 基金项目:国家自然科学基金资助项目(61310306028),浙江省自然 科学基金资助项目(Y15F020053)

更新日期/Last Update: 2016-01-20