Loading...

Improve Performance of Higher Order Statistics in Spatial and Frequency Domains in Blind Image Steganalysis

Shakeri, Ehsan | 2013

1521 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 44785 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Ghaemmaghami, Shahrokh
  7. Abstract:
  8. Blind image steganalysis is a technique used to, which require no prior information about the steganographic method applied to the stego im- age, determine whether the image contains an embedded message or not. The basic idea of blind steganalysis is to extract some features sensitive to information hiding, and then exploit classifiers for judging whether a given test image contains a secret message.The main focus of this research is to design an choose features sen-sitive to the embedding changes. In fact, we use high order moments in different domains, such as spatial, DCT and multi-resolution do-main, in order to improve the performance of existing steganalyzers.Accordingly, First, we propose an efficient blind image steganalysis method based on high order moments of singular values of contourlet coefficients of image and image noise. Analytical and experimental results reveal advantage of the proposed method over its counterpart steganalyzers. Next, another scheme is developed based on zernike moments and contourlet transform. Experimental results show that the proposed features are highly sensitive to the change made by the embedding process. Finally, a scheme is developed based on combin-ing the statistics of zernike moments with 274-D feature set of Pevny that yield about 6% higher stego detection accuracy, as compared to three baseline steganalyzers. Using proposed scheme, in addition toexcellent performance for weak steganography methods, the average accuracy of our scheme against secure steganography schemes such as PQ and NSF5 is satisfactory
  9. Keywords:
  10. Contourlet Transform ; Eigenvalue Decomposition ; Singular Value Decomposition (SVD) ; Zernike Moment ; Discrete Cosine Transform ; Steganography ; Blind Watermarking

 Digital Object List

 Bookmark

...see more