Loading...

Steganalysis Using Statistical Properties of Digital Signal

Khosravirad, Saeed Reza | 2009

698 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 39745 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Ghaemmaghami, Shahrokh; Eghlidos, Taraneh
  7. Abstract:
  8. Steganography is the art and technique of concealing secret message in ordinary data cover, transmitted over a public channel, in a way that eavesdroppers, as well as the channel users, cannot detect the presence of the secret message. However, steganalysis tries to detect this type of covert communication using some effective analysis techniques. Steganalysis is often based on statistical properties of the suspicious signal that are expected to change due to the message embedding process. Secret message (which mostly is a set of pseudo-random bits because of cryptography) affects the statistical features of the cover signal. So far, many steganalysis techniques have been reported that are generally designed based on alterations supposed to occur in statistical properties of the signal. A new scheme for steganalysis of the LSB embedding is proposed in this thesis based on a novel idea of Closure of Sets (CoS). The presented scheme comes with a broad range of applications, in both senses of the signal type and the embedding domain, as long as the message is steganographied using the random or sequential LSB embedding method. The proposed scheme is used for steganalysis of the LSB Steganography of grayscale TIFF and JPEG images and audio signals, employing a set of accurate and monotone features which are extracted based on the CoS definition. It is shown that significant improvement to the detection accuracy in steganalysis is achieved using the proposed scheme, particularly at low embedding rates. The proposed steganalysis scheme has been shown to be a reliable system that outperforms the state-of-the-art LSB steganalysis methods, in terms of detection accuracy and false alarm probability.

  9. Keywords:
  10. Steganography ; Steganalysis ; Least Significant Bit ; Signal Statistical Properties

 Digital Object List

 Bookmark