From e1c34dc36c5b5d8e9470b0baa3b5e5f80a8647e9 Mon Sep 17 00:00:00 2001 From: couchot Date: Fri, 5 Oct 2012 16:41:21 +0200 Subject: [PATCH] mineur --- analysis.tex | 9 ----- biblio.bib | 101 ++++++++++++++++++++++++++++++++++++++++++++++++ intro.tex | 28 ++++++++++---- main.tex | 72 ++++------------------------------ ourapproach.tex | 18 ++------- stc.tex | 5 +-- 6 files changed, 134 insertions(+), 99 deletions(-) diff --git a/analysis.tex b/analysis.tex index 4ae4244..2af26fb 100644 --- a/analysis.tex +++ b/analysis.tex @@ -33,12 +33,3 @@ scheme can avoid the LSB replacement style asymmetry, and thus it should make the detection slightly more difficult than the LSBM approach based on our experiments -random LSB selection: coarse, easily tractable, easilly detactable - - - -feature-based : precise but time consuming - -security need : message has to be encrypted before beeing introduced. Using efficient and secured cryptographic approach - -LSBR: to coarse (apply hugo breaker on it) diff --git a/biblio.bib b/biblio.bib index 47c0c00..2ddab01 100644 --- a/biblio.bib +++ b/biblio.bib @@ -374,3 +374,104 @@ author = {Jessica J. Fridrich and OPTannote = {} } + + + + +@article{Liu:2008:FMP:1284917.1285196, + author = {Liu, Qingzhong and Sung, Andrew H. and Chen, Zhongxue and Xu, Jianyun}, + title = {Feature mining and pattern classification for steganalysis of LSB matching steganography in grayscale images}, + journal = {Pattern Recogn.}, + issue_date = {January, 2008}, + volume = {41}, + number = {1}, + month = jan, + year = {2008}, + issn = {0031-3203}, + pages = {56--66}, + numpages = {11}, + url = {http://dx.doi.org/10.1016/j.patcog.2007.06.005}, + doi = {10.1016/j.patcog.2007.06.005}, + acmid = {1285196}, + publisher = {Elsevier Science Inc.}, + address = {New York, NY, USA}, + keywords = {DENFIS, Image complexity, LSB matching, SVMRFE, Steganalysis}, +} + + + +@InProceedings{LHS08, + author = {Bin Li and Jiwu Huang and Yun Q. Shi}, + title = {Textural features based universal steganalysis}, + OPTcrossref = {}, + OPTkey = {}, + booktitle = {Proc. SPIE 6819}, + pages = {12}, + year = {2008}, + OPTeditor = {}, + OPTvolume = {6819}, + + OPTseries = {}, + OPTaddress = {}, + month = feb, + OPTorganization = {}, + OPTpublisher = {}, + OPTnote = {}, + OPTannote = {} +} + + +@InProceedings{KF11, + author = {Jan Kodovský and Jessica Fridrich}, + title = {Steganalysis in high dimensions: Fusing classifiers built on random subspaces}, + OPTcrossref = {}, + OPTkey = {}, + booktitle = { Proc. SPIE, Electronic Imaging, Media Watermarking, Security, and Forensics XIII}, + OPTpages = {}, + year = {2011}, + OPTeditor = {}, + OPTvolume = {}, + OPTnumber = {}, + OPTseries = {}, + OPTaddress = {}, + OPTmonth = {}, + OPTorganization = {}, + OPTpublisher = {}, + OPTnote = {}, + OPTannote = {} +} + +@article{DBLP:journals/tifs/KodovskyFH12, + author = {Jan Kodovsk{\'y} and + Jessica J. Fridrich and + Vojtech Holub}, + title = {Ensemble Classifiers for Steganalysis of Digital Media}, + journal = {IEEE Transactions on Information Forensics and Security}, + volume = {7}, + number = {2}, + year = {2012}, + pages = {432-444}, + ee = {http://dx.doi.org/10.1109/TIFS.2011.2175919}, + bibsource = {DBLP, http://dblp.uni-trier.de} +} + + + +@article{Fillatre:2012:ASL:2333143.2333587, + author = {Fillatre, Lionel}, + title = {Adaptive Steganalysis of Least Significant Bit Replacement in Grayscale Natural Images}, + journal = {Trans. Sig. Proc.}, + issue_date = {February 2012}, + volume = {60}, + number = {2}, + month = feb, + year = {2012}, + issn = {1053-587X}, + pages = {556--569}, + numpages = {14}, + url = {http://dx.doi.org/10.1109/TSP.2011.2174231}, + doi = {10.1109/TSP.2011.2174231}, + acmid = {2333587}, + publisher = {IEEE Press}, + address = {Piscataway, NJ, USA}, +} \ No newline at end of file diff --git a/intro.tex b/intro.tex index f8974ce..dd24570 100644 --- a/intro.tex +++ b/intro.tex @@ -16,13 +16,27 @@ cannot be applied there to discover stego-images in LSBM. The most accurate detectors for this matching are universal steganalysers such as~\cite{LHS08,DBLP:conf/ih/2005,FK12} which classify images thanks to extracted features from neighboring elements of noise residual. + LSB matching revisited (LSBMR)~\cite{Mielikainen06} have been recently introduced. -This scheme deals with pairs of pixels instead of individual ones. -It thus allows to decrease the number of modified bits per cover pixel -for the same payload compared to LSB replacement and LSBM and -and avoids the LSB replacement style asymmetry. Unfortunately, -detectors referenced above are able to distinguish between -stego content images and cover images. +For a given pair of pixels, in which the LSB +of the first pixel carries one bit of secret message, and the relationship +(odd–even combination) of the two pixel values carries +another bit of secret message. + + + +In such a way, the modification +rate of pixels can decrease from 0.5 to 0.375 bits/pixel +(bpp) in the case of a maximum embedding rate, meaning fewer +changes to the cover image at the same payload compared to +LSB replacement and LSBM. It is also shown that such a new +scheme can avoid the LSB replacement style asymmetry, and +thus it should make the detection slightly more difficult than the +LSBM approach based on our experiments + + + + Instead of (efficiently) modifying LSBs, there is also a need to select pixels whose value modification minimizes a distortion function. @@ -69,7 +83,7 @@ Section~\ref{sec:experiments} shows experiments on image quality, steganalytic e and compare them to state of the art steganographic schemes. Finally, concluding notes and future works are given in section~\ref{sec:concl} -theory : ? + diff --git a/main.tex b/main.tex index fcbcdbb..e68d1ca 100755 --- a/main.tex +++ b/main.tex @@ -1,4 +1,4 @@ -\documentclass[draft,journal]{IEEEtran} +\documentclass[journal]{IEEEtran} \usepackage{subfig} \usepackage{color} \usepackage{graphicx} @@ -8,20 +8,8 @@ \begin{document} -% -% paper title -% can use linebreaks \\ within to get better formatting as desired -\title{Bare Demo of IEEEtran.cls for Journals} -% -% -% author names and IEEE memberships -% note positions of commas and nonbreaking spaces ( ~ ) LaTeX will not break -% a structure at a ~ so this keeps an author's name from being broken across -% two lines. -% use \thanks{} to gain access to the first footnote area -% a separate \thanks must be used for each paragraph as LaTeX2e's \thanks -% was not built to handle multiple paragraphs -% +\title{STABYLO: STeganography with cAnny, Bbs, binarY embedding at LOw cost} + \author{Jean-Fran\c cois Couchot, Raphael Couturier, and Christophe Guyeux* FEMTO-ST Institute, UMR 6174 CNRS\\ @@ -33,13 +21,9 @@ } \newcommand{\JFC}[1]{\begin{color}{green}\textit{#1}\end{color}} % make the title area -%\maketitle +\maketitle -\begin{abstract} -%\boldmath -The abstract goes here. -\end{abstract} \begin{IEEEkeywords} %IEEEtran, journal, \LaTeX, paper, template. @@ -54,61 +38,21 @@ Steganography, least-significant-bit (LSB)-based steganography, edge detection, -\section{Introduction} +\section{Introduction}\label{sec:intro} \input{intro.tex} -\section{Analysis of Steganographic Approaches} -\input{analysis.tex} - -This work considers digital images as covers and fondation is -spatial least significant-bit (LSB) replacement. -I this data hiding scheme a subset of all the LSB of the cover image is modified -with a secret bit stream depending on to a key, the cover, and the message to embed. -This well studied steganographic approach never decreases (resp. increases) -pixel with even value (resp. odd value) and may break structural symetry. -This structure modification is detectable by statistical approaches -and thus by steganalysis methods~\cite{Dumitrescu:2005:LSB:1073170.1073176,DBLP:conf/ih/2005,FK12}. - - -random LSB selection: coarse, easily tractable, easilly detactable - - -feature-based : precise but time consuming - -security need : message has to be encrypted before beeing introduced. Using efficient and secured cryptographic approach - -LSBR: to coarse (apply hugo breaker on it) - -\section{Our Approach} +\section{Our Approach}\label{sec:ourapproach} \input{ourapproach.tex} -% Image Quality: Edge Image Steganography +\section{Experiments}\label{sec:experiments} -% Security aspect: -% BBS-based cprotographic version of the message -% Enlarging embeding efficiency: -% Syndrome treillis code -\input{stc.tex} -\section{Conclusion} +\section{Conclusion}\label{sec:concl} The conclusion goes here. -\appendices -\section{Proof of the First Zonklar Equation} -Appendix one text goes here. - -% you can choose not to have a title for an appendix -% if you want by leaving the argument blank -\section{} -Appendix two text goes here. - - -% use section* for acknowledgement -\section*{Acknowledgment} -The authors would like to thank... \bibliographystyle{plain} \bibliography{biblio} diff --git a/ourapproach.tex b/ourapproach.tex index 2bf476f..28f8263 100644 --- a/ourapproach.tex +++ b/ourapproach.tex @@ -33,22 +33,10 @@ and inside the extraction one(Fig.~\ref{fig:sch:ext}). \subsection{Steganalysis} -LSB : -"Adaptive steganalysis of Least Significant Bit replacement in grayscale natural images" -Structural LSB Detectors: -\verb+http://dde.binghamton.edu/download/structural_lsb_detectors/+ +Détailler \cite{Fillatre:2012:ASL:2333143.2333587} -Vainqueur du BOSS challenge - - -ensemble: - - -G. Gül and F. Kurugollu. A new methodology in steganalysis : Breaking highly -undetactable steganograpy (HUGO). In Information Hiding, 13th International -Workshop, volume 6958 of LNCS, pages 71–84, Prague, Czech Republic, May 18– -20, 2011. +Vainqueur du BOSS challenge~\cite{DBLP:journals/tifs/KodovskyFH12} \subsection{Data Embedding} @@ -60,7 +48,7 @@ Workshop, volume 6958 of LNCS, pages 71–84, Prague, Czech Republic, May 18– Image Quality: Edge Image Steganography \JFC{Raphael, les fuzzy edge detection sont souvent utilisés. il faudrait comparer les approches en terme de nombre de bits retournés, - en terme de complexité. } + en terme de complexité. } \cite{KF11} Presentation des algos de detection de contour diff --git a/stc.tex b/stc.tex index 6570c73..e8c695c 100644 --- a/stc.tex +++ b/stc.tex @@ -58,10 +58,7 @@ $2^n-1$ pixels needs $1-1/2^n$ average changes. -Unfortunately, - - -for any given $H$, finding $y$ that solves $Hy=m$ and that +Unfortunately, for any given $H$, finding $y$ that solves $Hy=m$ and that that minimizes $D_X(x,y)$ has exponential complexity with respect to $n$. The Syndrome-Trellis Codes (STC) presented by Filler et al. in~\cite{DBLP:conf/mediaforensics/FillerJF10} -- 2.39.5