diff --git a/Article/Fig_method/algoMulti.tex b/Article/Fig_method/algoMulti.tex
new file mode 100644
index 0000000000000000000000000000000000000000..a0054750e01b797bc7da4b01db9259387ad42412
--- /dev/null
+++ b/Article/Fig_method/algoMulti.tex
@@ -0,0 +1,43 @@
+\begin{algorithm}[h]
+ \SetAlgoLined
+ \SetKwInOut{Input}{input}
+ \SetKwInOut{Output}{output}
+
+ \SetKwData{lm}{LocMax}
+ \SetKwData{nullset}{$\emptyset$}
+ \SetKwData{ortho}{$\vec{AB}_\perp$}
+ \SetKwData{eps}{$\varepsilon_{ini}$}
+ \SetKwData{pta}{$A$}
+ \SetKwData{ptb}{$B$}
+ \SetKwData{Result}{Result}
+
+ \SetKwArray{occmask}{OccupancyMask}
+ \SetKwArray{bseg}{BlurredSegment}
+ \SetKwArray{bslist}{ListOfBlurredSegments}
+
+ \SetKwFunction{locmax}{ComputeAndSortGradientLocalMax}
+ \SetKwFunction{sortgradmax}{SortByGradientMagnitude}
+ \SetKwFunction{taille}{Size}
+ \SetKwFunction{cardinal}{Card}
+
+ \SetKwData{Begin}{Start}
+ \SetKwData{End}{End}
+
+ \Input{Stroke points \pta, \ptb}
+ \Output{\textit{\bslist} $\rightarrow$ list of detected blurred segments}
+ \BlankLine
+ \bslist $\leftarrow$ \nullset\;
+ \occmask $\leftarrow$ \nullset\;
+ \lm $\leftarrow$ \locmax (\pta, \ptb)\;
+ \BlankLine
+ \For{$i \leftarrow 0$ \KwTo \taille(\lm)}{
+ \bseg $\leftarrow$ detect (\lm[i], \ortho, \eps, \occmask)\;
+ \For{$j \leftarrow 0$ \KwTo \cardinal (\bseg)}{
+ \occmask $\leftarrow$ \bseg[j]\;
+ }
+ \bslist $\leftarrow$ \bseg\;
+ }
+
+ \caption{MultiDetection:
+ finds all blurred segments under an input stroke.}
+\end{algorithm}
diff --git a/Article/Fig_method/edgeDir.readme b/Article/Fig_method/edgeDir.readme
new file mode 100644
index 0000000000000000000000000000000000000000..b7923f481987ed07c4cc7397772a8e4f962205c1
--- /dev/null
+++ b/Article/Fig_method/edgeDir.readme
@@ -0,0 +1,5 @@
+Images obtenues pas le test testEdgeDir.txt
+suivi de la commande de detection multiple : Ctrl-M
+avec deux modalites pilotees par Ctrl-E
+
+Sur gimp selection de la zone (34,166)(444,314).
diff --git a/Article/Fig_method/multiEdge_capt.png b/Article/Fig_method/multiEdge_capt.png
new file mode 100644
index 0000000000000000000000000000000000000000..b4e3f6ca0ac98557b607e9f47db25c09860b86a3
Binary files /dev/null and b/Article/Fig_method/multiEdge_capt.png differ
diff --git a/Article/Fig_method/multiEdge_zoom.png b/Article/Fig_method/multiEdge_zoom.png
new file mode 100644
index 0000000000000000000000000000000000000000..ad0c1abee3914b3d161a321225ed19f6ae6ba03e
Binary files /dev/null and b/Article/Fig_method/multiEdge_zoom.png differ
diff --git a/Article/Fig_method/multiStroke_capt.png b/Article/Fig_method/multiStroke_capt.png
new file mode 100644
index 0000000000000000000000000000000000000000..083195576f14d01fbf0d2be9999a9a6301232b07
Binary files /dev/null and b/Article/Fig_method/multiStroke_capt.png differ
diff --git a/Article/Fig_method/multiStroke_zoom.png b/Article/Fig_method/multiStroke_zoom.png
new file mode 100644
index 0000000000000000000000000000000000000000..9c9f9261b1eb863a41eb6e291a399f984978c798
Binary files /dev/null and b/Article/Fig_method/multiStroke_zoom.png differ
diff --git a/Article/Fig_method/testEdgeDir.txt b/Article/Fig_method/testEdgeDir.txt
new file mode 100644
index 0000000000000000000000000000000000000000..7e3274e1defbf89e33c37c296bc3ab66c0a90d1a
--- /dev/null
+++ b/Article/Fig_method/testEdgeDir.txt
@@ -0,0 +1,2 @@
+287 142
+326 103
diff --git a/Article/Fig_method/testVoisin.txt b/Article/Fig_method/testVoisin.txt
new file mode 100644
index 0000000000000000000000000000000000000000..82d7ef2672a8f76f99255fe4dbca84beb65faf9b
--- /dev/null
+++ b/Article/Fig_method/testVoisin.txt
@@ -0,0 +1,2 @@
+329 210
+323 245
diff --git a/Article/Fig_method/voisin.readme b/Article/Fig_method/voisin.readme
new file mode 100644
index 0000000000000000000000000000000000000000..14eaff83963e65cefcc3f6a94e1915e161202d1f
--- /dev/null
+++ b/Article/Fig_method/voisin.readme
@@ -0,0 +1,4 @@
+Images obtenues pas le test testvoisin.txt
+suivi de la commande de detection de bords : Ctrl-E
+
+Sur gimp selection de la zone (282,90)(411,156).
diff --git a/Article/Fig_method/voisinGrad_capt.png b/Article/Fig_method/voisinGrad_capt.png
new file mode 100644
index 0000000000000000000000000000000000000000..602c946c76c80cffe1b584a8b5b1a1e749b39501
Binary files /dev/null and b/Article/Fig_method/voisinGrad_capt.png differ
diff --git a/Article/Fig_method/voisinGrad_zoom.png b/Article/Fig_method/voisinGrad_zoom.png
new file mode 100644
index 0000000000000000000000000000000000000000..c0523578542e2a6546d0316024368cf8f0552e25
Binary files /dev/null and b/Article/Fig_method/voisinGrad_zoom.png differ
diff --git a/Article/Fig_method/voisinImage_capt.png b/Article/Fig_method/voisinImage_capt.png
new file mode 100644
index 0000000000000000000000000000000000000000..b57890145fdcf06cf2a17969e305a754fef12820
Binary files /dev/null and b/Article/Fig_method/voisinImage_capt.png differ
diff --git a/Article/Fig_method/voisinImage_zoom.png b/Article/Fig_method/voisinImage_zoom.png
new file mode 100644
index 0000000000000000000000000000000000000000..7d198c97aa5c60d04863c8bbc1c9387ecc134e14
Binary files /dev/null and b/Article/Fig_method/voisinImage_zoom.png differ
diff --git a/Article/Fig_method/voisinMulti_capt.png b/Article/Fig_method/voisinMulti_capt.png
new file mode 100644
index 0000000000000000000000000000000000000000..9768fffe054c92f2a1992c743d01d5c5eb0303c6
Binary files /dev/null and b/Article/Fig_method/voisinMulti_capt.png differ
diff --git a/Article/Fig_method/voisinMulti_zoom.png b/Article/Fig_method/voisinMulti_zoom.png
new file mode 100644
index 0000000000000000000000000000000000000000..1f65858d139a536490e80c4bd356eec531d429ea
Binary files /dev/null and b/Article/Fig_method/voisinMulti_zoom.png differ
diff --git a/Article/Fig_method/voisinSingle_capt.png b/Article/Fig_method/voisinSingle_capt.png
new file mode 100644
index 0000000000000000000000000000000000000000..53eb072fa2582a64f03f2e4d0a45912226577544
Binary files /dev/null and b/Article/Fig_method/voisinSingle_capt.png differ
diff --git a/Article/Fig_method/voisinSingle_zoom.png b/Article/Fig_method/voisinSingle_zoom.png
new file mode 100644
index 0000000000000000000000000000000000000000..4b729e9888faf38d65d24c579f7bd0b64c3e6af0
Binary files /dev/null and b/Article/Fig_method/voisinSingle_zoom.png differ
diff --git a/Article/method.tex b/Article/method.tex
index c2ce3b9ad21a9fa5594c619810641c8a667234f6..63adc79987dea8f28c6cc62bb1fd97d46bada938 100755
--- a/Article/method.tex
+++ b/Article/method.tex
@@ -2,7 +2,7 @@
\subsection{Workflow of the detection process}
-The blurred segment detector workflow is summerized
+The work-flow of the blurred segment detection process is summerized
in the following figure.
\begin{figure}[h]
@@ -12,8 +12,8 @@ in the following figure.
\put(-2,18){\scriptsize $(A,B)$}
\put(-2,15){\vector(1,0){24}}
\put(24,0){\framebox(56,30)}
- \put(24,15){\makebox(56,10){Initial}}
- \put(24,3){\makebox(56,10){detection}}
+ \put(24,16){\makebox(56,10){Initial}}
+ \put(24,4){\makebox(56,10){detection}}
\put(86,18){\scriptsize $\mathcal{B}_{1}$}
\put(80,15){\vector(1,0){22}}
%\put(102,0){\framebox(56,30)}
@@ -25,15 +25,15 @@ in the following figure.
\put(159,18){\scriptsize $(C,\vec{D})$}
\put(158,15){\vector(1,0){28}}
\put(186,0){\framebox(56,30)}
- \put(186,15){\makebox(56,10){Fine}}
- \put(186,3){\makebox(60,10){tracking}}
+ \put(186,16){\makebox(56,10){Fine}}
+ \put(186,4){\makebox(60,10){tracking}}
\put(250,18){\scriptsize $\mathcal{B}_2$}
\put(242,15){\vector(1,0){24}}
\put(266,0){\framebox(56,30){Filtering}}
\put(330,18){\scriptsize $\mathcal{B}_3$}
\put(322,15){\vector(1,0){22}}
\end{picture}
- \caption{The detection method work flow.}
+ \caption{The detection process main work-flow.}
\label{fig:workflow}
\end{figure}
@@ -63,21 +63,108 @@ Setting $N=20$ shows a good behaviour on tested images.
The fine track output segment is finally filtered to remove artifacts
and outliers, and a solution blurred segment $\mathcal{B}_3$ is provided.
-\subsection{The thinning process}
+\subsection{Supervised blurred segment detection}
+
+In supervised context, the user draws an input stroke across the specific
+edge he wants to extract from the image.
+The detection method previously described is continuously run during mouse
+dragging and the output blurred segment is displayed on-the-fly.
+
+The method is very sensitive to the local conditions of the initial detection
+so that the output blurred segment may be quite unstable.
+In order to temper this undesirable behaviour for particular applications,
+the initial detection can be optionally run twice, the second fast scan being
+aligned on the first detection output.
+This strategy provides a first quick analysis of the local context before
+extracting the segment and contributes to notably stabilize the overall
+process.
+
+Another option, called multi-detection allows the detection of all the
+segments crossed by the input stroke $AB$.
+The multi-detection algorithm is displayed below.
+First the positions $M_j$ of the local maxima of the gradient magnitude found
+under the stroke are sorted from the highest to the lowest.
+For each of them the main detection process is run with three modifications:
+i) the initial detection takes $M_j$ and the orthogonal direction $AB_\perp$
+to the stroke as input to build a static scan of fixed width
+$\varepsilon_{ini}$, and $M_j$ is used as start point of the blurred segment;
+ii) a occupancy mask, initially empty, is filled in with the points of the
+detected blurred segments $\mathcal{B}_{j3}$ at the end of each successful
+detection;
+iii) points marked as occupied are rejected when selecting candidates for the
+blurred selegment extension in the fine tracking step.
+
+\input{Fig_method/algoMulti}
+%\input{algo1}
+
+Beyond the possible detection of a large set of edges at once, the
+multi-detection allows the detection of some unaccessible edges in
+classical single detection mode. This is particularly the case of edges
+that are quite close to a more salient edge with a higher gradient.
+The multi-detection detects both edges and the user may then select
+the awaited one.
-\subsection{Multi-detection}
+\begin{figure}[h]
+\center
+ \begin{tabular}{c@{\hspace{0.2cm}}c@{\hspace{0.2cm}}c@{\hspace{0.2cm}}c}
+ \includegraphics[width=0.22\textwidth]{Fig_method/voisinImage_zoom.png} &
+ \includegraphics[width=0.22\textwidth]{Fig_method/voisinGrad_zoom.png} &
+ \includegraphics[width=0.22\textwidth]{Fig_method/voisinSingle_zoom.png} &
+ \includegraphics[width=0.22\textwidth]{Fig_method/voisinMulti_zoom.png} \\
+ \parbox{0.22\textwidth}{\centering{\scriptsize{a)}}} &
+ \parbox{0.22\textwidth}{\centering{\scriptsize{b)}}} &
+ \parbox{0.22\textwidth}{\centering{\scriptsize{c)}}} &
+ \parbox{0.22\textwidth}{\centering{\scriptsize{d)}}}
+ \end{tabular}
+ \caption{Example of edge, detected only in multi-detection mode:
+ a) the stroke on the intensity image,
+ b) the gradient map,
+ c) the result of the classical single mode detection,
+ d) the result of the multi-detection. }
+ \label{fig:voisins}
+\end{figure}
+
+This detection procedure can be used to dectect as well straight edges
+as thin straight objects. In the first case, the gradient vectors of all
+edge points are assumed to be oriented in the same direction. But if the
+sign of the gradient direction is not considered, points with gradient in
+opposite directions are merged to build the same blurred segment, allowing
+the detection of both edges of a thin linear structure, like for instance
+the tile joins of \RefFigure{fig:edgeDir}.
+
+\begin{figure}[h]
+\center
+ \begin{tabular}{c@{\hspace{0.2cm}}c}
+ \includegraphics[width=0.46\textwidth]{Fig_method/multiStroke_zoom.png} &
+ \includegraphics[width=0.46\textwidth]{Fig_method/multiEdge_zoom.png} \\
+ \parbox{0.46\textwidth}{\centering{\scriptsize{a)
+ Detection of straight lines}}} &
+ \parbox{0.46\textwidth}{\centering{\scriptsize{b)
+ Detection of straight edges}}}
+ \end{tabular}
+ \caption{Testing the gradient direction to detect as well edges as
+ linear structures from the same input stroke.}
+ \label{fig:edgeDir}
+\end{figure}
+
+On that example, when a straight features detection is run
+(\RefFigure{fig:edgeDir} a)),
+a thick blurred segment which extends up to four tiles is provided.
+When a straight edge detection is run, a very thin blurred segment is
+built to follow only one join edge.
+The multi-detection can also be applied to both thin object or edge detection.
+In the latter case, the detection algorithm is run twice using opposite
+directions, so that in the exemple of figure (\RefFigure{fig:edgeDir} b)),
+both edges (in different colours) are highlighted.
+These two thin blurred segments are much shorter, probably because the
+tiles are not perfectly aligned.
+This example illustrates our detector versatility.
-Algo de multi-detection.
+\subsection{Automatic blurred segment detection}
-\subsection{Automatic detection}
+Explication et b\'etail de l'algo
\subsection{Implementation details}
A directional scanner is encoded as an iterator that provides successively
all the scan lines.
-
-Description de l'interface pour la d\'etection supervis\'ee d'un segment.
-
-Description de la d\'etection multiple.
-
-Description de la d\'etection automatique.
diff --git a/Article/notions.tex b/Article/notions.tex
index ee859f5aad1189b8f26bcb31807f5a615ea1b065..2d3368cc5fa625b6283b96df1c27194ba1fe2448 100755
--- a/Article/notions.tex
+++ b/Article/notions.tex
@@ -1,4 +1,4 @@
-\section{Base notions}
+\section{Theoretical background}
\subsection{Blurred segment}