Steganalysis of LSB replacement for multivariate Gaussian covers

Recently, some statistically optimal steganalyzers are proposed based on hypothesis testing theory, in which the cover pixels are supposed to be independent. However, the independent assumption is of limited interest since redundancy exists in natural images. In this paper, using a more appropriate image model considering pixel correlation, a new steganaly-sis method for the least significant bit replacement (LSB-R) steganography is proposed. First, the cover image is divided into non-overlapping equal-sized blocks and each block is modeled by a multivariate Gaussian distribution. Then, based on likelihood ratio test and formula derivation, a new detector of LSB-R is derived. The proposed detector is an extension of some previous works with improved detection performance.

[1]  Florent Retraint,et al.  A Cover Image Model For Reliable Steganalysis , 2011, Information Hiding.

[2]  Jessica J. Fridrich,et al.  Detecting LSB Steganography in Color and Gray-Scale Images , 2001, IEEE Multim..

[3]  Rainer Böhme,et al.  Revisiting weighted stego-image steganalysis , 2008, Electronic Imaging.

[4]  Jessica J. Fridrich,et al.  On estimation of secret message length in LSB steganography in spatial domain , 2004, IS&T/SPIE Electronic Imaging.

[5]  Lionel Fillatre.  Adaptive Steganalysis of Least Significant Bit Replacement in Grayscale Natural Images , 2012, IEEE Transactions on Signal Processing.

[6]  Rainer Böhme,et al.  Moving steganography and steganalysis from the laboratory into the real world , 2013, IH&MMSec '13.

[7]  Florent Retraint,et al.  An Asymptotically Uniformly Most Powerful Test for LSB Matching Detection , 2013, IEEE Transactions on Information Forensics and Security.

[8]  B. S. Manjunath,et al.  Detection of hiding in the least significant bit , 2004, IEEE Transactions on Signal Processing.

[9]  Florent Retraint,et al.  Statistical detection of information hiding based on adjacent pixels difference , 2012, 2012 Proceedings of the 20th European Signal Processing Conference (EUSIPCO).

[10]  Andrew D. Ker.  Fourth-order structural steganalysis and analysis of cover assumptions , 2006, Electronic Imaging.

[11]  Stéphane Mallat,et al.  Solving Inverse Problems With Piecewise Linear Estimators: From Gaussian Mixture Models to Structured Sparsity , 2010, IEEE Transactions on Image Processing.

[12]  Florent Retraint,et al.  Application of hypothesis testing theory for optimal detection of LSB matching data hiding , 2013, Signal Process..

[13]  Florent Retraint,et al.  Hidden information detection based on quantized Laplacian distribution , 2012, 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[14]  Florent Retraint,et al.  Hidden information detection using decision theory and quantized samples: Methodology, difficulties and results , 2014, Digit. Signal Process..

[15]  Yair Weiss,et al.  From learning models of natural image patches to whole image restoration , 2011, 2011 International Conference on Computer Vision.

[16]  Andrew D. Ker.  A General Framework for Structural Steganalysis of LSB Replacement , 2005, Information Hiding.

[17]  Sorina Dumitrescu,et al.  Detection of LSB steganography via sample pair analysis , 2002, IEEE Trans. Signal Process..

[18]  Tomás Pevný,et al.  "Break Our Steganographic System": The Ins and Outs of Organizing BOSS , 2011, Information Hiding.

[19]  Florent Retraint,et al.  Statistical Decision Methods in Hidden Information Detection , 2011, Information Hiding.