Merge pull request #89 from kostrykin/rfove

adds RFOVE tool

.gitignore

@@ -5,3 +5,5 @@ template.xml
 template_shed.yml
 .vscode
 /create_csvs.py
+.DS_Store
+._*

tools/rfove/.shed.yml

@@ -0,0 +1,8 @@
+categories:
+  - Imaging
+description: Perform segmentation region-based fitting of overlapping ellipses
+long_description: RFOVE is completely unsupervised, operates without any assumption or prior knowledge on the object’s shape and extends and improves the Decremental Ellipse Fitting Algorithm (DEFA). Both RFOVE and DEFA solve the multi-ellipse fitting problem by performing model selection that is guided by the minimization of the Akaike Information Criterion on a suitably defined shape complexity measure. However, in contrast to DEFA, RFOVE minimizes an objective function that allows for ellipses with higher degree of overlap and, thus, achieves better ellipse-based shape approximation.
+name: rfove
+owner: imgteam
+homepage_url: https://sites.google.com/site/costaspanagiotakis/research/cs
+remote_repository_url: https://github.com/BMCV/galaxy-image-analysis/tree/master/tools/rfove/

tools/rfove/rfove.xml

@@ -0,0 +1,52 @@
+<tool id="rfove" name="Perform segmentation region-based fitting of overlapping ellipses" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="20.05">
+    <description>with RFOVE</description>
+    <macros>
+        <token name="@TOOL_VERSION@">2023.11.12</token>
+        <token name="@VERSION_SUFFIX@">0</token>
+    </macros>
+    <edam_operations>
+        <edam_operation>operation_3443</edam_operation>
+    </edam_operations>
+    <xrefs>
+        <xref type="bio.tools">rfove</xref>
+    </xrefs>
+    <requirements> 
+        <container type="docker">docker.io/kostrykin/rfove:@TOOL_VERSION@</container>
+    </requirements>
+    <command detect_errors="aggressive">
+    <![CDATA[
+    #set $neighborhood_size = "%d" % ($half_neighborhood_size.value * 2 + 1)
+    /rfove
+        ${area_ub}
+        ${min_area_max_area_ratio_ub}
+        ${overlap_ub}
+        ${neighborhood_size}
+        '${dataset}'
+
+        output.tiff
+    ]]>
+    </command>
+    <inputs>
+        <param name="dataset" type="data" format="tiff,png" label="Dataset" />
+        <param name="area_ub" type="integer" label="Maximum ellipse area" min="5" value="250" />
+        <param name="min_area_max_area_ratio_ub" type="float" min="0" max="1" value="0.1" label="Upper bound of the ratio: minimum area / maximum area" />
+
+        <param name="overlap_ub" type="float" label="Maximum ellipse overlap" min="0" value="0.2" />
+        <param name="half_neighborhood_size" type="integer" label="Half neighborhood size" min="1" value="100" />
+    </inputs>
+    <outputs>
+        <data format="tiff" name="masks" from_work_dir="output.tiff" label="${tool.name} on ${on_string}" />
+    </outputs>
+    <tests>
+        <test>
+            <param name="dataset" value="BBBC033_C2_z28.png" />
+            <output name="masks" value="output.tiff" ftype="tiff" compare="sim_size" />
+        </test>
+    </tests>
+    <help>
+    RFOVE is completely unsupervised, operates without any assumption or prior knowledge on the object’s shape and extends and improves the Decremental Ellipse Fitting Algorithm (DEFA). Both RFOVE and DEFA solve the multi-ellipse fitting problem by performing model selection that is guided by the minimization of the Akaike Information Criterion on a suitably defined shape complexity measure. However, in contrast to DEFA, RFOVE minimizes an objective function that allows for ellipses with higher degree of overlap and, thus, achieves better ellipse-based shape approximation.
+    </help>
+    <citations>
+        <citation type="doi">10.1016/j.imavis.2019.09.001</citation>
+    </citations>
+</tool>