This section aims at justifying the lightweight attribute of our approach.
To be more precise, we compare the complexity of our schemes to some of
current state of the art of
-steganographic scheme, namely HUGO~\cite{DBLP:conf/ih/PevnyFB10},
+steganographic schemes, namely HUGO~\cite{DBLP:conf/ih/PevnyFB10},
WOW~\cite{conf/wifs/HolubF12}, and UNIWARD~\cite{HFD14}.
-Each of these scheme starts with the computation of the distortion cost
+Each of these schemes starts with the computation of the distortion cost
for each pixel switch and is later followed by the STC algorithm.
Since this last step is shared by all,
we separately evaluate this complexity.
-In all the rest of this section, we consider a $n \times n$ square image.
+In all the remainder of this section, we consider a $n \times n$ square image.
First of all, HUGO starts with computing the second order SPAM Features.
-This steps is in $\theta(n^2 + 2\times 343^2)$ due to the calculation
+This steps is in $\theta(n^2 + 2\times 343^2)$ due to the computation
of the difference arrays and next of the 686 features (of size 343).
Next for each pixel, the distortion measure is calculated by +1/-1 modifying
its value and computing again the SPAM
metrics, which is in $\theta(686)$. Computing the distance is thus in
$\theta(2\times 343^2)$ and this modification
is thus in $\theta(2\times 343^2 \times n^2)$.
-Ranking these results may be achieved with an insertion sort, which is in
-$2 \times n^2 \ln(n)$.
+Ranking these results may be achieved with a quick sort, which is in
+$\theta(2 \times n^2 \ln(n))$ for data of size $n^2$.
The overall complexity of the pixel selection is finally
$\theta(n^2 +2 \times 343^2 + 2\times 343^2 \times n^2 + 2 \times n^2 \ln(n))$, \textit{i.e},
$\theta(2 \times n^2(343^2 + \ln(n)))$.
What follows details the complexity of the distortion evaluation of the
UNIWARD scheme. This one is based to a convolution product $W$ of two elements
-of size $n$ and is again in $\theta(n^2 \times n^2\ln(n^2))$ and a sum $D$ of
+of size $n$ and is again in $\theta(n^2 \times n^2\ln(n^2))$,
+ and a sum $D$ of
these $W$ which is in $\theta(n^2)$.
This distortion computation step is thus in $\theta(6n^4\ln(n) + n^2)$.
-Our edge selection is based on a Canny Filter. When applied on a
+Our edge selection is based on a Canny filter. When applied on a
$n \times n$ square image, the noise reduction step is in $\theta(5^3 n^2)$.
-Next, let $T$ be the size of the canny mask.
+Next, let $T$ be the size of the Canny mask.
Computing gradients is in $\theta(4Tn^2)$ since derivatives of each direction (vertical or horizontal)
are in $\theta(2Tn^2)$.
Finally, thresholding with hysteresis is in $\theta(n^2)$.
\begin{center}
\includegraphics[scale=0.4]{complexity}
\end{center}
-\caption{Complexity evaluation of WOW/UNIWARD, HUGO, and STABYLO}
+\caption{Complexity evaluation of WOW/UNIWARD, HUGO, and STABYLO.}
\label{fig:compared}
\end{figure}
