Loading...

Universal image steganalysis against spatial-domain steganography based on energy distribution of singular values

Shojaei Hashemi, A ; Sharif University of Technology

513 Viewed
  1. Type of Document: Article
  2. Abstract:
  3. A passive image steganalysis method is proposed to universally detect spatial-domain steganography schemes. It is shown to have better performance than universal steganalyzers known to be powerful in spatial domain, including the WFLogSv and the WAM methods. This level of accuracy is the result of improving the WFLogSv steganalyzer by considering a more comprehensive relationship between the singular values of each image block and the linear correlation of the rows and the columns. That is, instead of the closeness of the lower singular values to zero, the energy distribution of the singular values is investigated. An innovative measure is proposed for this investigation, which is inspired from arithmetic mean-geometric mean inequality. Experimental results confirm the supremacy of the proposed steganalysis scheme over its counterparts
  4. Keywords:
  5. Arithmetic mean-geometric mean inequality ; Spatial-domain steganography ; Energy distributions ; Image blocks ; Linear correlation ; Singular values ; Spatial domains ; Steganalysis ; Electric power distribution ; Image analysis ; Information technology ; Singular value decomposition ; Steganography
  6. Source: 7th International Conference on Information Technology and Application, ICITA 2011 ; 2011 , Pages 179-183 ; 9780980326741 (ISBN)
  7. URL: http://www.icita.org/previous/icita2011/papers/ir-shojaei.pdf