Loading...

Investigating the Relation Between Pixel Cost, Steganographic Capacity and Statistical Detectability

Amiri, Rouhollah | 2014

629 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 46579 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Gholampour, Iman
  7. Abstract:
  8. Statistical detectability is a feature of a steganalyzer that states its power for distinguishing between cover and stego images. A steganographer tries to design the steganography algorithm such that the steganalyzer could not detect the stego images. As a result, designing an efficient steganography algorithm (assigning the pixel cost) that could minimize the statistical detectability is an important objective in steganography. However, making a sensible relation between the pixel cost and statistical detectability is an open problem.
    In this thesis, by modeling the steganalyzer using the LDA topic model, we analyze its error probability as a criterion of statistical detectability. Moreover, based on this model, we propose a new definition for capacity and analyze its relation with the pixel cost. After that, the relation between the statistical detectability and the pixel cost is investigated using a suitable steganography algorithm. The proposed steganographic scheam outperforms MG as the analytical based state-of-the-art algorithm. Besides, we define a factor called image cost for ranking images. According to experimental results, selecting proper cover images w.r.t. image cost may lead to better resisting in steganalysis. Finally, we propose a hypothetical model for group steganalyzer and analyze its performance in the group steganalysis structure
  9. Keywords:
  10. Watermarking Capacity ; Latent Dirrichlet Allocation (LDA) ; Pixel Cost ; Statistical Detectabilty

 Digital Object List

 Bookmark

No TOC