Loading...

Blind Universal Steganalysis in Multiple Actor Paradigms and its Relation to Pixel-Cost

Akhondi, Ali | 2014

862 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 46431 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Gholampour, Iman
  7. Abstract:
  8. Steganography is method for communicating confidential information through a non-trustworth in way which hides the existence of communication. For improving the security of steganography statistical detectability must decrease as such as possible. Despite the fact, that the quality of the relation between statistical detectability and amount of distortion engendered by embedding is still an open problem, problem of detectability reduces to problem of management of pixel embedding in order to minimization of distortion. As in wet paper coding methods, an optimum (or approximately optimum) algorithm proportioned to Pixel-cost has been offered, the current problem of steganography is to find a proper method to determine the Pixel-costs.
    Algorithms for specifying Pixel-cost apply a fitting distortion function an approximating Pixel-costs thereof. In this thesis instead of using a distortion function, we employ pooled steganalysis to determine Pixel-costs, with the intention to approximate steganography natural designing. Steganalysis is the attempt to discover steganography. In its classic paradigm, a given image is analyzed which, generally, is based on classification of images. In multi- actor paradigm, it is assumed multiple actor are active in sending pictures. Here finding the guilty actor is prominent. Actors are being clustered and the smallest cluster is taken as the guilty actor.
    In this thesis ,first we introduce a suitable pooled steganalyser for steganalysis of LSB matching based on rich model features is designed. We apply feature selection and feature extraction to overcome the curse dimensionality and also present a suitable distance function. Later we deal with comparison of image potential for embedding. In this part we present image-cost criteria. These criteria reveal randomness of a taken image. In last section we generalize image-cost to sub-image-cost. By comparison we arrived at a criteria for ranking of them. Then the concept, Pixel cost is consider as the average cost of the included sub-images in the main image. According to this concept, later, we apply steganography. Our experiments prove that our steganography algorithm enjoy high security and effectiveness especially in low embedding rate in juxtaposition with the MG algorithm.
  9. Keywords:
  10. Genetic Algorithm ; Batch Steganography ; Universal Steganalysis ; Distance Function ; Image-Cost ; Pixel Cost ; Subimage

 Digital Object List

 Bookmark