2 \subsection{Image Quality}
3 The visual quality of the STABYLO scheme is evaluated in this section.
4 Three metrics are computed in these experiments :
5 the Peak Signal to Noise Ratio (PSNR),
6 the PSNR-HVS-M~\cite{PSECAL07,psnrhvsm11} and the BIQI~\cite{MB10,biqi11}.
7 The first one is widely used but does not take into
8 account Human Visual System (HVS).
9 The two last ones have been designed to tackle this problem.
18 \subsection{Steganalysis}
22 The quality of our approach has been evaluated through the two
23 AUMP~\cite{Fillatre:2012:ASL:2333143.2333587}
24 and Ensemble Classifier~\cite{DBLP:journals/tifs/KodovskyFH12} based steganalysers.
25 Both aims at detecting hidden bits in grayscale natural images and are
26 considered as the state of the art of steganalysers in spatial domain~\cite{FK12}.
27 The former approach is based on a simplified parametric model of natural images.
28 Parameters are firstly estimated and a adaptive Asymptotically Uniformly Most Powerful
29 (AUMP) test is designed (theoretically and practically) to check whether
30 a natural image has stego content or not.
31 In the latter, the authors show that the
32 machine learning step, (which is often
33 implemented as support vector machine)
34 can be a favourably executed thanks to an Ensemble Classifiers.
38 \JFC{Raphael, il faut donner des résultats ici}