1 The flowcharts given in Fig.~\ref{fig:sch} summarize our steganography scheme denoted as to
2 STABYLO for STeganography with cAnny, Bbs, binarY embedding at LOw cost.
3 What follows successively details all the inner steps and flow inside
4 the embedding stage (Fig.\ref{fig:sch:emb})
5 and inside the extraction one (Fig.~\ref{fig:sch:ext}).
10 \subfloat[Data Embedding.]{
11 \begin{minipage}{0.49\textwidth}
13 \includegraphics[width=5cm]{emb.pdf}
14 %\includegraphics[width=5cm]{emb.ps}
19 \subfloat[Data Extraction.]{
20 \begin{minipage}{0.49\textwidth}
22 \includegraphics[width=5cm]{rec.pdf}
23 %\includegraphics[width=5cm]{rec.ps}
29 \caption{The STABYLO Scheme.}
36 \subsection{Data Embedding}
37 This section describes the main three steps of the STABYLO data embedding
42 \subsubsection{Edge Based Image Steganography}
45 Edge Based Image Steganography schemes
46 already studied~\cite{Luo:2010:EAI:1824719.1824720,DBLP:journals/eswa/ChenCL10,DBLP:conf/ih/PevnyFB10} differ
47 how they select edge pixels, and
48 how they modify these ones.
50 Image Quality: Edge Image Steganography
51 \JFC{Raphael, les fuzzy edge detection sont souvent utilisés.
52 il faudrait comparer les approches en terme de nombre de bits retournés,
53 en terme de complexité. } \cite{KF11}
54 \RC{Ben, à voir car on peut choisir le nombre de pixel avec canny. Supposons que les fuzzy edge soient retourne un peu plus de points, on sera probablement plus détectable... Finalement on devrait surement vendre notre truc en : on a choisi cet algo car il est performant en vitesse/qualité. Mais on peut aussi en utilisé d'autres :-)}
56 There are many techniques to detect edges in images. Main methods are filter
57 edge detection methods such as Sobel or Canny filter, low order methods such as
58 first order and second order methods, these methods are based on gradient or
59 Laplace operators and fuzzy edge methods which are based on fuzzy logic to
62 Of course, all the algorithms have advantages and drawbacks which depend on the
63 motivation to highlight edges. Unfortunately unless testing most of the
64 algorithms, which would require many times, it is quite difficult to have an
65 accurate idea on what would produce such algorithm compared to another. That is
66 why we have chosen canny algorithm which is well known, fast and implementable
67 on many kinds of architecture, such as FPGA, smartphone, desktop machines and
68 GPU. And of course, we do not pretend that this is the best solution.
71 First of all, let us discuss about compexity of edge detetction methods.
72 Let then $M$ and $N$ be the dimension of the original image.
73 According to~\cite{Hu:2007:HPE:1282866.1282944},
74 even if the fuzzy logic based edge detection methods~\cite{Tyan1993}
75 have promising results, its complexity is in $C_3 \times O(M \times N)$
76 whereas the complexity on the Canny method~\cite{Canny:1986:CAE:11274.11275}
77 is in $C_1 \times O(M \times N)$ where $C_1 < C_3$.
78 \JFC{Verifier ceci...}
79 In experiments detailled in this article, the canny method has been retained
80 but the whole approach can be updated to consider
81 the fuzzy logic edge detector.
83 Next, following~\cite{Luo:2010:EAI:1824719.1824720}, our scheme automatically
84 modifies canny parameters to get a sufficiently large set of edge bits: this
85 one is practically enlarged untill its size is at least twice as many larger
86 than the size of embedded message.
90 \subsubsection{Security Considerations}
91 Among methods of message encryption/decryption
92 (see~\cite{DBLP:journals/ejisec/FontaineG07} for a survey)
93 we implement the Blum-Goldwasser cryptosystem~\cite{Blum:1985:EPP:19478.19501}
94 which is based on the Blum Blum Shub~\cite{DBLP:conf/crypto/ShubBB82} Pseudo Random Number Generator (PRNG)
96 It has been indeed proven~\cite{DBLP:conf/crypto/ShubBB82} that this PRNG
97 has the cryptographically security property, \textit{i.e.},
98 for any sequence $L$ of output bits $x_i$, $x_{i+1}$, \ldots, $x_{i+L-1}$,
99 there is no algorithm, whose time complexity is polynomial in $L$, and
100 which allows to find $x_{i-1}$ and $x_{i+L}$ with a probability greater
102 Thus, even if the encrypted message would be extracted,
103 it would thus be not possible to retrieve the original one in a
111 \subsubsection{Minimizing Distortion with Syndrome-Treillis Codes}
115 \subsection{Data Extraction}
116 Message extraction summarized in Fig.~\ref{fig:sch:ext} follows data embedding
117 since there exists a reverse function for all its steps.
118 First of all, the same edge detection is applied to get set,
119 which is sufficiently large with respect to the message size given as a key.
120 Then the STC reverse algorithm is applied to retrieve the encrypted message.
121 Finally, the Blum-Goldwasser decryption function is executed and the original
122 message is extracted.