diff --git a/Makefile b/Makefile
index 36d1d3b..7afa0dd 100644
--- a/Makefile
+++ b/Makefile
@@ -13,15 +13,13 @@ PACKAGE_FILE_PATH=$(shell pwd)/lipase-$(LIPASE_VERSION).zip
 .PHONY: all
 all: test doc
 
-.PHONY:
-doc: doc/lipase.pdf
+.PHONY: doc
+doc:
+	cd doc; make 
 
-doc/lipase.pdf: doc/lipase.tex
-	cd doc; latexmk 
-
-.PHONY:
+.PHONY: clean_doc
 clean_doc:
-	cd doc; rm lipase.pdf lipase.aux lipase.fls lipase.log lipase.fdb_latexmk lipase.dvi; 
+	cd doc; make clean
 
 
 $(BUILD_ROOT_PATH)/jars/Lib/fr.univ-rennes1.ipr.lipase.lib.jar: $(LIB_SRC_FILES)
@@ -32,7 +30,7 @@ $(BUILD_ROOT_PATH)/jars/Lib/fr.univ-rennes1.ipr.lipase.lib.jar: $(LIB_SRC_FILES)
 		popd
 
 .PHONY: install_ij_opencv
-install_ij_opencv: 
+install_ij_opencv:
    # enable IJ-OpenCV-plugins site because it's disabled by default
 
     # graffy@graffy-ws2:~$ zcat /home/graffy/soft/Fiji.app/db.xml.gz | grep IJ-Op | grep -v "jar" | grep -v "<plu"
diff --git a/doc/Makefile b/doc/Makefile
new file mode 100644
index 0000000..d4b029d
--- /dev/null
+++ b/doc/Makefile
@@ -0,0 +1,29 @@
+RAW_IMAGES_ROOT?=/home/graffy/work/lipase/raw-images
+lipase.pdf: lipase.tex graphics
+	pdflatex lipase.tex
+
+# command used to convert traps sequences in the visible into png with maximum contrast
+TRAPS_VISIBLE_CONVERTER=convert -contrast-stretch 2%x1% -resize 25%
+
+GENERATED_GRAPHICS=\
+	graphics/res_soleil2018_GGH_GGH_2018_cin2_phiG_I_327_vis_-40_1_Pos0_img_000000000_DM300_nofilter_vis_000.png \
+	graphics/res_soleil2018_GGH_GGH_2018_cin2_phiG_I_327_vis_-40_1_Pos0_img_000000019_DM300_nofilter_vis_000.png \
+	graphics/res_soleil2018_GGH_GGH_2018_cin2_phiG_I_327_vis_-40_1_Pos0_img_000000039_DM300_nofilter_vis_000.png
+
+.PHONY: graphics
+graphics: $(GENERATED_GRAPHICS)
+
+graphics/res_soleil2018_GGH_GGH_2018_cin2_phiG_I_327_vis_-40_1_Pos0_img_000000000_DM300_nofilter_vis_000.png : $(RAW_IMAGES_ROOT)/res_soleil2018/GGH/GGH_2018_cin2_phiG_I_327_vis_-40_1/Pos0/img_000000000_DM300_nofilter_vis_000.tif
+	$(TRAPS_VISIBLE_CONVERTER) "$<" "$@"
+
+graphics/res_soleil2018_GGH_GGH_2018_cin2_phiG_I_327_vis_-40_1_Pos0_img_000000019_DM300_nofilter_vis_000.png : $(RAW_IMAGES_ROOT)/res_soleil2018/GGH/GGH_2018_cin2_phiG_I_327_vis_-40_1/Pos0/img_000000019_DM300_nofilter_vis_000.tif
+	$(TRAPS_VISIBLE_CONVERTER) "$<" "$@"
+
+
+graphics/res_soleil2018_GGH_GGH_2018_cin2_phiG_I_327_vis_-40_1_Pos0_img_000000039_DM300_nofilter_vis_000.png : $(RAW_IMAGES_ROOT)/res_soleil2018/GGH/GGH_2018_cin2_phiG_I_327_vis_-40_1/Pos0/img_000000039_DM300_nofilter_vis_000.tif
+	$(TRAPS_VISIBLE_CONVERTER) "$<" "$@"
+
+.PHONY: clean
+clean:
+	rm -f $(GENERATED_GRAPHICS)
+	rm -f lipase.pdf lipase.aux lipase.fls lipase.log lipase.fdb_latexmk lipase.dvi; 
diff --git a/doc/graphics/readme.md b/doc/graphics/readme.md
new file mode 100644
index 0000000..be4f8ec
--- /dev/null
+++ b/doc/graphics/readme.md
@@ -0,0 +1 @@
+the purpose of this file is to make sure the empty graphics directory can be part of the git repository (workaround to the fact that git doesn't version empty directories)
\ No newline at end of file
diff --git a/doc/lipase.tex b/doc/lipase.tex
index 0452b09..93da392 100644
--- a/doc/lipase.tex
+++ b/doc/lipase.tex
@@ -1,14 +1,26 @@
-\documentclass[a4paper]{article}
+\documentclass[a4paper, 10pt]{article}
 \usepackage[utf8]{inputenc}
+\usepackage{graphicx}
+\usepackage{subcaption}
+\usepackage[htt]{hyphenat} % allow hyphen inside texttt to avoid overfull hbox warnings
+\usepackage[english, french]{babel}
+\usepackage[margin=0.5in]{geometry} % default margins are too big for my taste: too much wasted space http://kb.mit.edu/confluence/pages/viewpage.action?pageId=3907057
+\usepackage{amsmath} % provides underset
+
+\hyphenation{tu-yau}
+
+
 \title{lipase}
 \author{Guillaume Raffy \and Véronique Vié }
 \begin{document}
+    \selectlanguage{english}
 
     \maketitle
 
     \section{catalog images}
 
         image prefix :
+        \selectlanguage{french}
         \begin{description}
             \item[AF]
             \item[blé] coupes de blé
@@ -45,5 +57,46 @@
             \item[cin2] autre échantillon similaire à cin1
             \item[cond5678] condition non réalistes
         \end{description}
+        \selectlanguage{english}
+
+    \section{computing background image for trap sequences}
+
+        Trap sequences show traps at fixed positions with particles that move over time, as shown in figure \ref{fig:trap_sequence1}. In order to detect the particles, we can subtract from each image a background image, which is an image of the scene without any particle.
+
+        If we suppose that particles are moving fast enough, we can estimate this background image $B$, as :
+
+        \begin{equation}
+            B(x,y) = \underset{t\in {1 \ldots T_{max}}}{\mathrm{median}} \{I(x,y,t)\} 
+        \end{equation}
+        where $I(x,y,t)$ is the value of the input sequence at time $t$ and on pixel position $(x,y)$ and $T_{max}$ is the number of frames in the sequence.
+
+        \begin{figure}
+            \centering
+            \begin{subfigure}[b]{0.3\textwidth}
+                \includegraphics[width=1.0\textwidth]{graphics/res_soleil2018_GGH_GGH_2018_cin2_phiG_I_327_vis_-40_1_Pos0_img_000000000_DM300_nofilter_vis_000.png}
+                %\includegraphics[width=\textwidth]{1.png}
+                \caption{Frame 0}
+                %\label{fig:1}
+            \end{subfigure}
+            ~
+            \begin{subfigure}[b]{0.3\textwidth}
+                \includegraphics[width=1.0\textwidth]{graphics/res_soleil2018_GGH_GGH_2018_cin2_phiG_I_327_vis_-40_1_Pos0_img_000000019_DM300_nofilter_vis_000.png}
+                %\includegraphics[width=\textwidth]{1.png}
+                \caption{Frame 19}
+                %\label{fig:1}
+            \end{subfigure}
+            ~
+            \begin{subfigure}[b]{0.3\textwidth}
+                \includegraphics[width=1.0\textwidth]{graphics/res_soleil2018_GGH_GGH_2018_cin2_phiG_I_327_vis_-40_1_Pos0_img_000000039_DM300_nofilter_vis_000.png}
+                %\includegraphics[width=\textwidth]{1.png}
+                \caption{Frame 39}
+                %\label{fig:1}
+            \end{subfigure}
+
+
+            \caption{Example of trap sequence (\texttt{res\_soleil2018/GGH/GGH\_2018\_cin2\_phiG\_I\_327\_vis\_-40\_1/Pos0})}
+            \label{fig:trap_sequence1}
+        \end{figure}
+        
 
 \end{document}
\ No newline at end of file
diff --git a/doc/notes/journal.md b/doc/notes/journal.md
new file mode 100644
index 0000000..33fe0a7
--- /dev/null
+++ b/doc/notes/journal.md
@@ -0,0 +1,19 @@
+## 12/03/2020
+
+- graffy : investigations on how to compute white for sequences containing traps:
+    - `telemos.WhiteEstimator` is not suitable for a sequence containing traps, as the resulting white image displays. `telemos.WhiteEstimator` is expecting the input images to only contain small particles.
+    - supposing that particles move in the sequences containing trap, the white image can be computed by 
+
+
+## 17/03/2020
+
+- graffy
+    - I made some manual tests for extracting background of `res_soleil2018/GGH/GGH_2018_cin2_phiG_I_327_vis_-40_1/Pos0`. For this sequence, the median image provides a worse background image than the last frame of the sequence, which contains nearly no particle. So for now we'll use the last frame as a background for the moment, but we have to find a better background estimator than just the median image (an idea is for example to keep for each pixel the value that is the most frequent in the sequence)
+    - Some manual tests on the sequence `res_soleil2018/GGH/GGH_2018_cin2_phiG_I_327_vis_-40_1/Pos0` showed that subtracting the background image makes particles more obviusly visible, as expected. Applying a threshold allows to isolate the particles, but estimating the surface ofd particles from that will be tricky, as only the border of particles have a value which is very different from the background. Of course a fill hole operator could fill rthe particles but this can be fragile as the borders of particles are not always clearly visible on all its surrounding.
+    - a more promising solution for detecting particles would be to compute for each pixel :
+    	- the radial profile around the pixel
+    	- the radial variance profile around the pixel
+    	then, particle are the pixels for which
+    		- the radial variance is low (this means that there's a radial symmetry around the pixel)
+    		- and the radial profile shows a peak. The position of the peak gives the radius of the particle
+    	- In order to efficiently compute these radial profile and radial variance profiles on a whole image, convolutions methods can provide dramatic speedups. This technique has already been used in https://subversion.ipr.univ-rennes1.fr/repos/main/projects/antipode (`texori.py`). Unfortunately, I can't reuse this code as it makes extensive used of `numpy` (and imagej's jython doesn't support `numpy`)