Additionally to (efficiently) modifying LSBs, there is also a need to select pixels whose value
modification minimizes a distortion function.
-This distortion may be computed thanks to feature vectors that are embedded for instance in the steganalysers
+This distortion may be computed thanks to feature vectors that
+ are embedded for instance in the steganalysers
referenced above.
-The Highly Undetectable steGO (HUGO) method~\cite{DBLP:conf/ih/PevnyFB10} is one of the most efficient instance of such a scheme.
-It takes into account so-called SPAM features
-%(whose size is larger than $10^7$)
-to avoid over-fitting a particular
-steganalyser. Thus a distortion measure for each pixel is individually determined as the sum of the differences between
+The Highly Undetectable steGO (HUGO) method~\cite{DBLP:conf/ih/PevnyFB10},
+WOW~\cite{conf/wifs/HolubF12}, and UNIWARD~\cite{HFD14}
+are some of the most efficient instances of such a scheme.
+
+HUGO takes into account so-called SPAM features.
+Thus a distortion measure for each
+pixel is individually determined as the sum of the differences between
the features of the SPAM computed from the cover and from the stego images.
-Due to this features set, HUGO allows to embed messages that are $7$ times longer than the former ones with the same level of
+The features embedded in WOW and UNIWARD are based on Wavelet-based
+directional filter. Thus, similarelly, the distortion function is
+the sum of the differences between these wavelet coefficients
+computed from the cover and from the stego images.
+
+
+Due to this distortion measures, HUGO, WOW and UNIWARD allow
+to embed messages that are $7$ times longer than the former
+ones with the same level of
indetectability as LSB matching.
-However, this improvement is time consuming, mainly due to the distortion function
-computation.
+However, this improvement has a larger computation cost, mainly due to
+ the distortion function
+calculus.
There remains a large place between random selection of LSB and feature based modification of pixel values.
Section~\ref{sec:ourapproach} presents the details of the proposed steganographic scheme and applies it on a running example. Among its technical description,
its adaptive aspect is emphasized.
Section~\ref{sub:complexity} presents the overall complexity of our approach
-and compares it to the HUGO's one.
+and compares it to HUGO, WOW, and UNIWARD.
Section~\ref{sec:experiments} shows experiments on image quality, steganalysis evaluation, and compares them to the state of the art steganographic schemes.
Finally, concluding notes and future work are given in Section~\ref{sec:concl}.
