JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE) ›› 2011, Vol. 41 ›› Issue (2): 75-79.

• Articles • Previous Articles     Next Articles

A contourlet transform-based steganalysis algorithm

SHI Xiayang, WANG Yufei, HU Yongjian   

  1. School of Electronic and Information Engineering, South China University of Technology,
     Guangzhou 510641, China
  • Received:2010-11-04 Online:2011-04-16 Published:2010-11-04

Abstract:

 A universal steganalysis method was proposed by using the superior property of the contourlet transform with representation of an image. It merged the highorder statistics model of coefficient moments statistics, noise residual moments statistics, and characteristic function moments in the high frequency subband of the contourlet domain. At the same time, a nonlinear support vector machine(SVM) classifier was used to classify JSteg, Jphide, F5 and Outguess with different embedding rates. Experimental results showed that the proposed method has the superion discriminative performance  for  most of steganography methods. Compared with the classical wavelet, the contourlet transform has better detection effect to capture  slight differences during embedding messages.
 

Key words: steganalysis, contourlet transform, statistics characteristic

[1] ZHAO Wen-zhong. Self-adaptive multisensor image fusion algorithm based on dual-tree complex wavelet-Contourlet transform [J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2010, 40(4): 144-148.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!