DOC: Fill in missing ITK Filter help text. Improves generated python …

…documentation (#707)

* Fill in missing parts of ITK filter documentation
* Fill in missing help text for SurfaceNets filter
* Only generate docs for plugins that were compiled. Reduces print statement noise.

---------

Signed-off-by: Michael Jackson <[email protected]>

src/Plugins/ComplexCore/src/ComplexCore/Filters/SurfaceNetsFilter.cpp

@@ -58,7 +58,7 @@ Parameters SurfaceNetsFilter::parameters() const
 {
   Parameters params;
   params.insertSeparator(Parameters::Separator{"Input Parameters"});
-  params.insertLinkableParameter(std::make_unique<BoolParameter>(k_ApplySmoothing_Key, "Apply smoothing operations", "", false));
+  params.insertLinkableParameter(std::make_unique<BoolParameter>(k_ApplySmoothing_Key, "Apply smoothing operations", "Use the built in smmothing operation.", false));
 
   params.insertSeparator(Parameters::Separator{"Smoothing Values"});
   params.insert(std::make_unique<Int32Parameter>(k_SmoothingIterations_Key, "Relaxation Iterations", "Number of relaxation iterations to perform. More iterations causes more smoothing.", 20));

src/Plugins/ITKImageProcessing/docs/ITKBinaryContourImage.md

@@ -26,7 +26,7 @@ https://www.insight-journal.org/browse/publication/217
 
 | Name | Type | Description |
 |------|------|-------------|
-| FullyConnected | bool | Set/Get whether the connected components are defined strictly by face connectivity or by face+edge+vertex connectivity. Default is FullyConnectedOff. For objects that are 1 pixel wide, use FullyConnectedOn. |
+| FullyConnected | bool | Whether the connected components are defined strictly by face connectivity (False) or by face+edge+vertex connectivity (True). Default is False |
 | BackgroundValue | float64 | Set/Get the background value used to mark the pixels not on the border of the objects. |
 | ForegroundValue | float64 | Set/Get the foreground value used to identify the objects in the input and output images. |
 

src/Plugins/ITKImageProcessing/docs/ITKBinaryDilateImage.md

@@ -29,9 +29,9 @@ for arbitrary size and shape". IEEE Transactions on Image Processing. Vol. 9. No
 |------|------|-------------|
 | KernelRadius | uint32 | Set the radius of the kernel structuring element. |
 | KernelType | KernelEnum | Set the kernel or structuring element used for the morphology. |
-| BackgroundValue | float64 |  |
-| ForegroundValue | float64 |  |
-| BoundaryToForeground | bool |  |
+| BackgroundValue | float64 | The eroded pixels will receive the BackgroundValue. Default = non positive minimum  |
+| ForegroundValue | float64 | The pixel value considered 'Foreground' that will be eroded  |
+| BoundaryToForeground | bool | "Mark the boundary between foreground and background. |
 
 ## Required Geometry
 

src/Plugins/ITKImageProcessing/docs/ITKBinaryErodeImage.md

@@ -29,9 +29,9 @@ for arbitrary size and shape". IEEE Transactions on Image Processing. Vol. 9. No
 |------|------|-------------|
 | KernelRadius | uint32 | Set the radius of the kernel structuring element. |
 | KernelType | KernelEnum | Set the kernel or structuring element used for the morphology. |
-| BackgroundValue | float64 |  |
-| ForegroundValue | float64 |  |
-| BoundaryToForeground | bool |  |
+| BackgroundValue | float64 | The eroded pixels will receive the BackgroundValue. Default = non positive minimum  |
+| ForegroundValue | float64 | The pixel value considered 'Foreground' that will be eroded  |
+| BoundaryToForeground | bool | "Mark the boundary between foreground and background. |
 
 ## Required Geometry
 

src/Plugins/ITKImageProcessing/docs/ITKBinaryOpeningByReconstructionImage.md

@@ -30,7 +30,7 @@ This implementation was taken from the Insight Journal paper: https://www.insigh
 | KernelType | KernelEnum | Set the kernel or structuring element used for the morphology. |
 | ForegroundValue | float64 | Set the value in the image to consider as "foreground". Defaults to maximum value of PixelType. |
 | BackgroundValue | float64 | Set the value in eroded part of the image. Defaults to zero |
-| FullyConnected | bool | Set/Get whether the connected components are defined strictly by face connectivity or by face+edge+vertex connectivity. Default is FullyConnectedOff. For objects that are 1 pixel wide, use FullyConnectedOn. |
+| FullyConnected | bool | Whether the connected components are defined strictly by face connectivity (False) or by face+edge+vertex connectivity (True). Default is False |
 
 ## Required Geometry
 

src/Plugins/ITKImageProcessing/docs/ITKBinaryProjectionImage.md

@@ -8,6 +8,9 @@ ITKImageStatistics (ImageStatistics)
 
 ## Description
 
+Image projection is a very common task in image analysis to reduce the dimension of an image. A base
+filter is provided with some specialized filters which implement different projection methods.
+
 This class was contributed to the Insight Journal by Gaetan Lehmann. The original paper can be found at https://www.insight-journal.org/browse/publication/71 
 
 ### Author
@@ -29,7 +32,7 @@ This class was contributed to the Insight Journal by Gaetan Lehmann. The origina
 
 | Name | Type | Description |
 |------|------|-------------|
-| ProjectionDimension | uint32 |  |
+| ProjectionDimension | uint32 | The index of the projection dimension |
 | ForegroundValue | float64 | Set the value in the image to consider as "foreground". Defaults to maximum value of PixelType. Subclasses may alias this to DilateValue or ErodeValue. |
 | BackgroundValue | float64 | Set the value used as "background". Any pixel value which is not DilateValue is considered background. BackgroundValue is used for defining boundary conditions. Defaults to NumericTraits<PixelType>::NonpositiveMin() . |
 

src/Plugins/ITKImageProcessing/docs/ITKBinaryThresholdImage.md

@@ -18,12 +18,13 @@ The filter expect both images to have the same number of dimensions.
 
 The default values for LowerThreshold and UpperThreshold are: LowerThreshold = NumericTraits<TInput>::NonpositiveMin() ; UpperThreshold = NumericTraits<TInput>::max() ; Therefore, generally only one of these needs to be set, depending on whether the user wants to threshold above or below the desired threshold.
 
+Set the thresholds. The default lower threshold is NumericTraits<InputPixelType>::NonpositiveMin() . The default upper threshold is NumericTraits<InputPixelType>::max . An exception is thrown if the lower threshold is greater than the upper threshold.
 ## Parameters
 
 | Name | Type | Description |
 |------|------|-------------|
-| LowerThreshold | float64 |  |
-| UpperThreshold | float64 | Set the thresholds. The default lower threshold is NumericTraits<InputPixelType>::NonpositiveMin() . The default upper threshold is NumericTraits<InputPixelType>::max . An exception is thrown if the lower threshold is greater than the upper threshold. |
+| LowerThreshold | float64 | The lower threshold that a pixel value could be and still be considered 'Inside Value' |
+| UpperThreshold | float64 | The upper threshold that a pixel value could be and still be considered 'Inside Value'|
 | InsideValue | uint8 | Set the "inside" pixel value. The default value NumericTraits<OutputPixelType>::max() |
 | OutsideValue | uint8 | Set the "outside" pixel value. The default value NumericTraits<OutputPixelType>::ZeroValue() . |
 

src/Plugins/ITKImageProcessing/docs/ITKClosingByReconstructionImage.md

@@ -31,7 +31,7 @@ Applications", Second Edition, Springer, 2003.
 |------|------|-------------|
 | KernelRadius | uint32 | Set the radius of the kernel structuring element. |
 | KernelType | KernelEnum | Set the kernel or structuring element used for the morphology. |
-| FullyConnected | bool | Set/Get whether the connected components are defined strictly by face connectivity or by face+edge+vertex connectivity. Default is FullyConnectedOff. For objects that are 1 pixel wide, use FullyConnectedOn. |
+| FullyConnected | bool | Whether the connected components are defined strictly by face connectivity (False) or by face+edge+vertex connectivity (True). Default is False |
 | PreserveIntensities | bool | Set/Get whether the original intensities of the image retained for those pixels unaffected by the opening by reconstruction. If Off, the output pixel contrast will be reduced. |
 
 ## Required Geometry

src/Plugins/ITKImageProcessing/docs/ITKDilateObjectMorphologyImage.md

@@ -19,7 +19,7 @@ If a pixel's value is equal to the object value and the pixel is adjacent to a n
 |------|------|-------------|
 | KernelRadius | uint32 | Set the radius of the kernel structuring element. |
 | KernelType | KernelEnum | Set the kernel or structuring element used for the morphology. |
-| ObjectValue | float64 |  |
+| ObjectValue | float64 | The pixel value of the 'Object' to be dilated |
 
 ## Required Geometry
 

src/Plugins/ITKImageProcessing/docs/ITKErodeObjectMorphologyImage.md

@@ -19,7 +19,7 @@ If the pixel covered by the center of the kernel has the pixel value ObjectValue
 |------|------|-------------|
 | KernelRadius | uint32 | Set the radius of the kernel structuring element. |
 | KernelType | KernelEnum | Set the kernel or structuring element used for the morphology. |
-| ObjectValue | float64 |  |
+| ObjectValue | float64 | The pixel value of the 'Object' to be eroded |
 | BackgroundValue | float64 | Set the value to be assigned to eroded pixels |
 
 ## Required Geometry

src/Plugins/ITKImageProcessing/docs/ITKGrayscaleFillholeImage.md

@@ -22,7 +22,7 @@ Principles and Applications", Second Edition, Springer, 2003.* ReconstructionByE
 
 | Name | Type | Description |
 |------|------|-------------|
-| FullyConnected | bool | Set/Get whether the connected components are defined strictly by face connectivity or by face+edge+vertex connectivity. Default is FullyConnectedOff. For objects that are 1 pixel wide, use FullyConnectedOn. |
+| FullyConnected | bool | Whether the connected components are defined strictly by face connectivity (False) or by face+edge+vertex connectivity (True). Default is False |
 
 ## Required Geometry
 

src/Plugins/ITKImageProcessing/docs/ITKGrayscaleGrindPeakImage.md

@@ -24,7 +24,7 @@ Principles and Applications", Second Edition, Springer, 2003.* GrayscaleGeodesic
 
 | Name | Type | Description |
 |------|------|-------------|
-| FullyConnected | bool | Set/Get whether the connected components are defined strictly by face connectivity or by face+edge+vertex connectivity. Default is FullyConnectedOff. For objects that are 1 pixel wide, use FullyConnectedOn. |
+| FullyConnected | bool | Whether the connected components are defined strictly by face connectivity (False) or by face+edge+vertex connectivity (True). Default is False |
 
 ## Required Geometry
 

src/Plugins/ITKImageProcessing/docs/ITKHConvexImage.md

@@ -21,7 +21,7 @@ Principles and Applications", Second Edition, Springer, 2003.* GrayscaleGeodesic
 | Name | Type | Description |
 |------|------|-------------|
 | Height | float64 | Set/Get the height that a local maximum must be above the local background (local contrast) in order to survive the processing. Local maxima below this value are replaced with an estimate of the local background. |
-| FullyConnected | bool | Set/Get whether the connected components are defined strictly by face connectivity or by face+edge+vertex connectivity. Default is FullyConnectedOff. For objects that are 1 pixel wide, use FullyConnectedOn. |
+| FullyConnected | bool | Whether the connected components are defined strictly by face connectivity (False) or by face+edge+vertex connectivity (True). Default is False |
 
 ## Required Geometry
 

src/Plugins/ITKImageProcessing/docs/ITKHMinimaImage.md

@@ -23,7 +23,7 @@ Principles and Applications", Second Edition, Springer, 2003.* GrayscaleGeodesic
 | Name | Type | Description |
 |------|------|-------------|
 | Height | float64 | Set/Get the height that a local maximum must be above the local background (local contrast) in order to survive the processing. Local maxima below this value are replaced with an estimate of the local background. |
-| FullyConnected | bool | Set/Get whether the connected components are defined strictly by face connectivity or by face+edge+vertex connectivity. Default is FullyConnectedOff. For objects that are 1 pixel wide, use FullyConnectedOn. |
+| FullyConnected | bool | Whether the connected components are defined strictly by face connectivity (False) or by face+edge+vertex connectivity (True). Default is False |
 
 ## Required Geometry
 

src/Plugins/ITKImageProcessing/docs/ITKLabelContourImage.md

@@ -26,7 +26,7 @@ https://www.insight-journal.org/browse/publication/217
 
 | Name | Type | Description |
 |------|------|-------------|
-| FullyConnected | bool | Set/Get whether the connected components are defined strictly by face connectivity or by face+edge+vertex connectivity. Default is FullyConnectedOff. *Note* For objects that are 1 pixel wide, use FullyConnectedOn. |
+| FullyConnected | bool | Whether the connected components are defined strictly by face connectivity (False) or by face+edge+vertex connectivity (True). Default is False |
 | BackgroundValue | float64 | Set/Get the background value used to identify the objects and mark the pixels not on the border of the objects. |
 
 ## Required Geometry

src/Plugins/ITKImageProcessing/docs/ITKMorphologicalWatershedImage.md

@@ -27,9 +27,9 @@ This code was contributed in the Insight Journal paper: "The watershed transform
 
 | Name | Type | Description |
 |------|------|-------------|
-| Level | float64 |  |
+| Level | float64 | Set the 'level' variable to the filter |
 | MarkWatershedLine | bool | Set/Get whether the watershed pixel must be marked or not. Default is true. Set it to false do not only avoid writing watershed pixels, it also decrease algorithm complexity. |
-| FullyConnected | bool | Set/Get whether the connected components are defined strictly by face connectivity or by face+edge+vertex connectivity. Default is FullyConnectedOff. For objects that are 1 pixel wide, use FullyConnectedOn. |
+| FullyConnected | bool | Whether the connected components are defined strictly by face connectivity (False) or by face+edge+vertex connectivity (True). Default is False |
 
 ## Required Geometry
 

src/Plugins/ITKImageProcessing/docs/ITKOpeningByReconstructionImage.md

@@ -31,7 +31,7 @@ Applications", Second Edition, Springer, 2003.
 |------|------|-------------|
 | KernelRadius | uint32 | Set the radius of the kernel structuring element. |
 | KernelType | KernelEnum | Set the kernel or structuring element used for the morphology. |
-| FullyConnected | bool | Set/Get whether the connected components are defined strictly by face connectivity or by face+edge+vertex connectivity. Default is FullyConnectedOff. For objects that are 1 pixel wide, use FullyConnectedOn. |
+| FullyConnected | bool | Whether the connected components are defined strictly by face connectivity (False) or by face+edge+vertex connectivity (True). Default is False |
 | PreserveIntensities | bool | Set/Get whether the original intensities of the image retained for those pixels unaffected by the opening by reconstruction. If Off, the output pixel contrast will be reduced. |
 
 ## Required Geometry

src/Plugins/ITKImageProcessing/docs/ITKValuedRegionalMaximaImage.md

@@ -2,6 +2,14 @@
 
 Transforms the image so that any pixel that is not a regional maxima is set to the minimum value for the pixel type. Pixels that are regional maxima retain their value.
 
+The pixel neighborhood is controlled by the FullyConnected attribute. All adjacent pixels are included in
+the neighborhood when FullyConnected=True and the diagonally adjacent pixels are not included when
+FullyConnected=False. Different terminology is often used to describe neighborhoods – one common ex-
+ample is the “connectivity” notation, which refers to the number of pixels in the neighborhood. FullyCon-
+nected=False corresponds to a connectivty of 4 in 2D and 6 in 3D, while FullyConnected=True corresponds
+to a connectivity of 8 in 2D and 26 in 3D. FullyConnected=False is also commonly referred to as “face
+connected”.
+
 ## Group (Subgroup)
 
 ITKMathematicalMorphology (MathematicalMorphology)
@@ -26,7 +34,7 @@ This code was contributed in the Insight Journal paper: "Finding regional extrem
 
 | Name | Type | Description |
 |------|------|-------------|
-| FullyConnected | bool |  |
+| FullyConnected | bool | Whether the connected components are defined strictly by face connectivity (False) or by face+edge+vertex connectivity (True). Default is False |
 
 ## Required Geometry
 

src/Plugins/ITKImageProcessing/docs/ITKValuedRegionalMinimaImage.md

@@ -25,7 +25,7 @@ This code was contributed in the Insight Journal paper: "Finding regional extrem
 
 | Name | Type | Description |
 |------|------|-------------|
-| FullyConnected | bool |  |
+| FullyConnected | bool | Whether the connected components are defined strictly by face connectivity (False) or by face+edge+vertex connectivity (True). Default is False |
 
 ## Required Geometry
 

src/Plugins/ITKImageProcessing/src/ITKImageProcessing/Filters/ITKBinaryContourImage.cpp

@@ -73,9 +73,9 @@ Parameters ITKBinaryContourImage::parameters() const
 {
   Parameters params;
   params.insertSeparator(Parameters::Separator{"Input Parameters"});
-  params.insert(std::make_unique<BoolParameter>(k_FullyConnected_Key, "FullyConnected",
-                                                "Set/Get whether the connected components are defined strictly by face connectivity or by face+edge+vertex connectivity. Default is FullyConnectedOff. "
-                                                "For objects that are 1 pixel wide, use FullyConnectedOn.",
+  params.insert(std::make_unique<BoolParameter>(k_FullyConnected_Key, "Fully Connected Components",
+                                                "Whether the connected components are defined strictly by face connectivity (False) or by face+edge+vertex connectivity (True). Default is False"
+                                                "For objects that are 1 pixel wide, use True.",
                                                 false));
   params.insert(std::make_unique<Float64Parameter>(k_BackgroundValue_Key, "BackgroundValue", "Set/Get the background value used to mark the pixels not on the border of the objects.", 0.0));
   params.insert(std::make_unique<Float64Parameter>(k_ForegroundValue_Key, "ForegroundValue", "Set/Get the foreground value used to identify the objects in the input and output images.", 1.0));

src/Plugins/ITKImageProcessing/src/ITKImageProcessing/Filters/ITKBinaryDilateImage.cpp

@@ -83,9 +83,9 @@ Parameters ITKBinaryDilateImage::parameters() const
                                                           std::vector<std::string>{"X", "Y", "Z"}));
   params.insert(std::make_unique<ChoicesParameter>(k_KernelType_Key, "KernelType", "The shape of the kernel to use. 0=Annulas, 1=Ball, 2=Box, 3=Cross", static_cast<uint64>(itk::simple::sitkBall),
                                                    ChoicesParameter::Choices{"Annulus", "Ball", "Box", "Cross"}));
-  params.insert(std::make_unique<Float64Parameter>(k_BackgroundValue_Key, "BackgroundValue", "The background value of the image", 0.0));
-  params.insert(std::make_unique<Float64Parameter>(k_ForegroundValue_Key, "ForegroundValue", "The foreground value of the image", 1.0));
-  params.insert(std::make_unique<BoolParameter>(k_BoundaryToForeground_Key, "BoundaryToForeground", "", false));
+  params.insert(std::make_unique<Float64Parameter>(k_BackgroundValue_Key, "BackgroundValue", "The eroded pixels will receive the BackgroundValue. Default = non positive minimum", 0.0));
+  params.insert(std::make_unique<Float64Parameter>(k_ForegroundValue_Key, "ForegroundValue", "The pixel value considered 'Foreground' that will be eroded ", 1.0));
+  params.insert(std::make_unique<BoolParameter>(k_BoundaryToForeground_Key, "BoundaryToForeground", "Mark the boundary between foreground and background.", false));
 
   params.insertSeparator(Parameters::Separator{"Required Input Cell Data"});
   params.insert(std::make_unique<GeometrySelectionParameter>(k_SelectedImageGeomPath_Key, "Image Geometry", "Select the Image Geometry Group from the DataStructure.", DataPath({"Image Geometry"}),

src/Plugins/ITKImageProcessing/src/ITKImageProcessing/Filters/ITKBinaryErodeImage.cpp

@@ -83,9 +83,9 @@ Parameters ITKBinaryErodeImage::parameters() const
                                                           std::vector<std::string>{"X", "Y", "Z"}));
   params.insert(std::make_unique<ChoicesParameter>(k_KernelType_Key, "KernelType", "The shape of the kernel to use. 0=Annulas, 1=Ball, 2=Box, 3=Cross", static_cast<uint64>(itk::simple::sitkBall),
                                                    ChoicesParameter::Choices{"Annulus", "Ball", "Box", "Cross"}));
-  params.insert(std::make_unique<Float64Parameter>(k_BackgroundValue_Key, "BackgroundValue", "The background value of the image", 0.0));
-  params.insert(std::make_unique<Float64Parameter>(k_ForegroundValue_Key, "ForegroundValue", "The foreground value of the image", 1.0));
-  params.insert(std::make_unique<BoolParameter>(k_BoundaryToForeground_Key, "BoundaryToForeground", "", true));
+  params.insert(std::make_unique<Float64Parameter>(k_BackgroundValue_Key, "BackgroundValue", "The eroded pixels will receive the BackgroundValue. Default = non positive minimum", 0.0));
+  params.insert(std::make_unique<Float64Parameter>(k_ForegroundValue_Key, "ForegroundValue", "The pixel value considered 'Foreground' that will be eroded ", 1.0));
+  params.insert(std::make_unique<BoolParameter>(k_BoundaryToForeground_Key, "BoundaryToForeground", "Mark the boundary between foreground and background.", true));
 
   params.insertSeparator(Parameters::Separator{"Required Input Cell Data"});
   params.insert(std::make_unique<GeometrySelectionParameter>(k_SelectedImageGeomPath_Key, "Image Geometry", "Select the Image Geometry Group from the DataStructure.", DataPath({"Image Geometry"}),