Represents a model (i.e. a model with cells and scalar values) and their common operations. Also allows batch operations on collection of models. More...

#include <milxModel.h>

Inheritance diagram for milx::Model:

Public Member Functions
	Model ()
	Standard constructor.

	Model (vtkSmartPointer< vtkPolyData > model)
	Constructor that copies the input model.

virtual	~Model ()
	Standard Destructor.

void	AddInput (vtkSmartPointer< vtkPolyData > model)
	Assigns and coninually appends meshes provided to the class.

void	AddArray (vtkSmartPointer< vtkDataArray > array)
	Appends an array to the model.

void	SetInput (vtkSmartPointer< vtkPolyData > model)
	Assigns the input model to class. More...

void	SetInputData (vtkSmartPointer< vtkPolyData > model)
	Assigns the input model to class. Same as SetInput and meant for VTK version > 5.

void	SetTransform (vtkSmartPointer< vtkAbstractTransform > newTransform)
	Transforms the model by given transform.

void	SetGraph (vtkSmartPointer< vtkMutableUndirectedGraph > graph)
	Converts a graph to polydata/mesh ready for display.

void	SetPoints (vtkSmartPointer< vtkPoints > modelPoints)
	Assign points to the model.

void	SetPolys (vtkSmartPointer< vtkCellArray > modelPolys)
	Assign polygons to the model via the array given.

void	SetVectors (vtkSmartPointer< vtkDataArray > modelVectors)
	Sets the vector field of the mesh (vectors per vertex of the surface)

void	SetScalars (vtkSmartPointer< vtkDataArray > modelScalars)
	Sets the scalar field of the mesh (values per vertex of the surface)

void	SetActiveScalars (std::string nameOfArray)
	Sets the array of name as the scalar field of the mesh (values per vertex of the surface)

vtkSmartPointer< vtkPolyData > &	Result ()
	Returns the current model, i.e. the result of the latest operation. More...

vtkSmartPointer< vtkPolyData > &	PreviousResult ()
	Returns the previous model, i.e. the result of the penultimate operation. More...

vtkSmartPointer< vtkPolyData > &	GetInput ()
	Returns the input model, i.e. the very most initial model. More...

vtkSmartPointer< vtkPolyData > &	GetOutput ()
	Returns the current model, i.e. the result of the latest operation ITK/VTK style.

vtkAlgorithmOutput *	GetProducerPort ()
	Returns the current model, i.e. the result of the latest operation ITK/VTK style.

vtkIdType	GetNumberOfPoints ()
	Returns the total number of points of the model currently held.

vtkIdType	GetNumberOfArrays ()
	Returns the total number of point-based arrays in the model.

vtkSmartPointer< vtkTransform > &	GetTransform ()
	Returns the current net transform applied to the model, i.e. concatenations of all (rigid/linear) transforms.

vtkSmartPointer< vtkPoints >	GetPoints ()
	Returns the points of the model.

vtkSmartPointer< vtkDataArray >	GetScalars ()
	Returns the active scalars of the model.

void	GetScalarRange (double *range)
	Returns the scalar range of the model.

vtkSmartPointer< vtkDataArray >	GetVectors ()
	Returns the active vectors of the model.

vtkSmartPointer< vtkDataArray >	GetNormals ()
	Returns the normal vectors of the model.

vtkSmartPointer< vtkDataArray >	GetTensors ()
	Returns the active tensors (3x3 matrix per vertex) of the model.

void	Update ()
	Updates the model to be the most current.

void	Clean ()
	Removes duplicate points etc. from model. More...

void	Triangulate ()
	Ensures the polygons in the model are triangulated. Useful when some filters required triangulation before use. More...

void	MatchInformation (vtkSmartPointer< vtkPolyData > matchMesh, const bool rescale)
	Matches the centroid and centroid scale of the model to the one given.

void	IterativeClosestPointsRegistration (vtkSmartPointer< vtkPolyData > fixedMesh, const bool similarity=false, const int maxPoints=1024)
	Computes the ICP registration of the current mesh to the one provided.

void	LandmarkBasedRegistration (vtkSmartPointer< vtkPolyData > fixedMesh, const bool similarity=false)
	Computes the landmark transform registration of the current mesh to the one provided. Requires point correspondence and same number of points.

void	Decimate (const float reduceFactor)
	Decimate the mesh (in terms of points) using the Decimate PRO. Generally use QuadricDecimate() instead. More...

void	QuadricDecimate (const float reduceFactor)
	Decimate the mesh (in terms of points) using the Quadric algorithm. More...

void	ClusterDecimate ()
	Decimate the mesh (in terms of points) using the Quadric Clustering algorithm. More...

void	LaplacianSmoothing (const int iterations)
	Smooth the mesh (in terms of points) using the Laplacian algorithm. It is unstable, use WindowedSincSmoothing() instead unless massive non-edge preserving smoothing is required.

void	WindowedSincSmoothing (const int iterations, const float passBand=0.1)
	Smooth the mesh (in terms of points) using the Taubin (1995) optimal filter.

void	Curvature (const bool meanCurvature=true)
	Compute the curvature (using Laplace-Beltrami operator) of the mesh.

void	Gradient ()
	Compute the gradient (using Laplace-Beltrami operator) of the scalars on the mesh.

void	DistanceMap (vtkSmartPointer< vtkPoints > loopPoints)
	Compute the distance map of the mesh points given a loop (as points) assumed to be on the mesh surface.

void	GaussianWeightScalars (float stddev, float clampValue)
	Evaluate the Gaussian function with the given standard deviation for all values of the scalars on the mesh. More...

void	Rotate (const bool xAxis, const bool yAxis, const bool zAxis, const float angle, const coordinate centre)
	Rotate the mesh along the selected axes from the centre by angle. More...

void	Flip (const bool xAxis, const bool yAxis, const bool zAxis)
	Flip the mesh along the selected axes. More...

void	Clip (const coordinateType belowValue, const coordinateType aboveValue)
	Clip the mesh based on a scalar threshold of the mesh scalars. More...

void	Threshold (const coordinateType belowVal, const coordinateType aboveVal, const coordinateType outsideVal)
	Threshold the scalars on the mesh based on a levels provided.

void	Threshold (const coordinateType belowVal, const coordinateType aboveVal, const coordinateType insideVal, const coordinateType outsideVal)
	Threshold the scalars on the mesh based on a levels provided. Inside value means a binary threshold is effectively done.

void	Mask (vtkSmartPointer< vtkPolyData > maskMesh)
	Mask the scalars on the mesh based on values of maskMesh being >= 1.

void	IsoSurface (vtkSmartPointer< vtkImageData > img, double minValue)

vtkSmartPointer< vtkImageData >	Voxelise (const unsigned char insideValue, double spacing, double bounds=NULL, const size_t padVoxels=1)

void	DelaunayGraph3D ()
	Compute the Delaunay 3D graph of the points in the mesh. More...

void	Delaunay2DTriangulation ()
	Compute the Delaunay 2D triangluation of the points in the mesh. More...

void	DelaunayTriangulation ()
	Compute the Delaunay 3D triangluation of the points in the mesh.

void	Append (vtkSmartPointer< vtkPolyData > model)
	Appends the model provided to the member into the current model.

void	Scale (const coordinateType scale, const bool useCentroid=true)

void	GenerateVertices ()
	Generate (only) vertices for display from point data. Good for when only points in mesh.

void	GenerateVertexScalars ()
	Generate vertex scalars for display from point data. This colours the mesh by point ids.

void	GenerateVerticesAs2DGlyphs (const GlyphType glyphType)
	Generate (only) vertices (as 2D glyphs like crosses etc.) for display from point data. Good for when only points in mesh.

void	GenerateTubes ()
	Generate tubes along edges of the mesh.

void	GeneratePointModel (const double newScale)
	Generate spheres of scaling for each point of the mesh.

void	GenerateSpheres (const double newScale)
	Generate spheres of scaling for each point of the mesh. Same as GeneratePointModel().

void	GenerateSampledPoints (const double distance)
	Generate sampled points (only points) for the mesh at given spacing on triangles.

void	GenerateVectorField (const double newScale=0.0, const bool useNormals=false)
	Generate vector field for vectors present (as an array) in the mesh at given scaling. More...

void	GenerateTensorField (const double newScale=0.0)
	Generate tensor field for tensors present (as an array) in the mesh at given scaling. More...

void	GenerateHedgehog (const double newScale=0.0)
	Generate a line field for vectors present (as an array) in the mesh at given scaling. More...

void	GenerateNormals (const int pointNormals=0)
	Generate normal vectors for each point on the mesh. If pointNormals is (0,1,2), then normals are for (points,cells,both).

void	GenerateKMeansClustering (int numberOfClusters)
	Generate k-means clustering for the point positions in the mesh.

void	GenerateQuantisedPoints (float quantiseFactor)
	Generate points of the mesh to have integral cordinates.

void	GenerateCappedBoundaries ()
	Generate capped boundaries for open meshes. This closes off open ends of a mesh.

void	GenerateRegions ()
	Generate connected regions for the mesh labelled by scalar values.

void	GenerateElevation (double x, double y, double z)
	Generate scalar field for the mesh proportional to the elevation (dot product) with a given vector/direction.

void	GenerateElevation ()
	Generate scalar field for the mesh proportional to the elevation in the z-direction.

void	ResetScalars ()
	Resets the scalars to zero of the current model. More...

void	FillScalars (const coordinateType value)
	Sets all the scalars of the current model to value. More...

void	RemoveScalars ()
	Removes all the scalars from the current model. Same as ClearScalars()

void	RemoveArrays ()
	Removes all the arrays from the current model. Same as ClearArrayss()

void	ConcatenateCollection (vtkSmartPointer< vtkPolyDataCollection > collection)
	Concatenates all the models in the collection together into one model. More...

void	ScaleCollection (vtkSmartPointer< vtkPolyDataCollection > collection, const coordinateType scale)
	Scales the coordinates of each point in all the models in the collection. More...

void	SmoothCollection (vtkSmartPointer< vtkPolyDataCollection > collection, const size_t iterations)
	Smoothes all points in all the models in the collection using the Windowed Sinc algorithm. More...

void	LaplacianCollection (vtkSmartPointer< vtkPolyDataCollection > collection, const size_t iterations)
	Smoothes all points in all the models in the collection using the Laplacian algorithm. More...

void	FlipCollection (vtkSmartPointer< vtkPolyDataCollection > collection, const bool xAxis, const bool yAxis, const bool zAxis)
	Flips all points in all the models in the collection using the axis provided. More...

void	DecimateCollection (vtkSmartPointer< vtkPolyDataCollection > collection, const coordinateType factor)
	Decimates all points in all the models in the collection using the Quadric algorithm algorithm. More...

void	SplitCollection (vtkSmartPointer< vtkPolyDataCollection > collection, vtkSmartPointer< vtkPolyDataCollection > components, std::vector< vtkSmartPointer< vtkPolyDataCollection > > &splitCollections)
	Splits each of the models in the collection into a collection of models. More...

void	ScalarDifferenceCollection (vtkSmartPointer< vtkPolyDataCollection > collection)
	Computes the difference in scalars cummulatively over the collection wrt the first mesh assuming the meshes have the same number of points. More...

void	ScalarDifferenceCollection (vtkSmartPointer< vtkPolyDataCollection > collection1, vtkSmartPointer< vtkPolyDataCollection > collection2, vtkSmartPointer< vtkPolyDataCollection > &results)

void	ScalarThresholdCollection (vtkSmartPointer< vtkPolyDataCollection > collection, const coordinateType aboveVal, const coordinateType belowVal)
	Computes the Threshold of scalar values of meshes in place over the collection wrt the first mesh assuming the meshes have the same number of points. More...

void	ClipCollection (vtkSmartPointer< vtkPolyDataCollection > collection, const coordinateType aboveVal, const coordinateType belowVal)
	Computes the Clipping of models via the Threshold of scalar values in place over the collection. More...

void	ScalarStatisticsCollection (vtkSmartPointer< vtkPolyDataCollection > collection)
	Computes the statistics (min, max, variance etc.) of the scalars over the collection assuming the meshes have the same number of points. More...

void	ScalarRemoveCollection (vtkSmartPointer< vtkPolyDataCollection > collection)
	Removes the scalars over the collection assuming the meshes have the same number of points. More...

void	ScalarCopyCollection (vtkSmartPointer< vtkPolyDataCollection > collection)
	Copies the scalars over the collection (from the first mesh) assuming the meshes have the same number of points. More...

double	MeanSquaredErrorCollection (vtkSmartPointer< vtkPolyDataCollection > collection)
	Computes the MSE of all models in collection accummulatively.

vtkSmartPointer< vtkPolyDataCollection >	ProcrustesAlignCollection (vtkSmartPointer< vtkPolyDataCollection > collection, bool rigid=false)
	Aligns the collection to the mean mesh (computed internally) of the collection assuming the meshes have the same number of points and correspondence. More...

vtkSmartPointer< vtkPolyDataCollection >	IterativeClosestPointsAlignCollection (vtkSmartPointer< vtkPolyDataCollection > collection, bool rigid=false, vtkSmartPointer< vtkTransformCollection > tCollection=0)
	Aligns the collection to the mean mesh (computed internally) of the collection without needing correspondence. More...

Static Public Member Functions
static void	Scale (vtkSmartPointer< vtkPolyData > model, const coordinateType scale, const bool useCentroid=true)
	Scales the coordinates of each point in the model provided by scale. More...

static void	ClearScalars (vtkSmartPointer< vtkPolyData > &mesh)
	Removes all the scalars from the current model. More...

static void	ClearArrays (vtkSmartPointer< vtkPolyData > &mesh)
	Removes all the arrays from the current model. More...

static void	ScalarDifference (vtkSmartPointer< vtkPolyData > model1, vtkSmartPointer< vtkPolyData > model2, vtkSmartPointer< vtkPolyData > result)
	Calculates differences within scalars of two meshes assuming the meshes have the same number of points. More...

static void	ThresholdScalars (vtkSmartPointer< vtkPolyData > model, const coordinateType aboveVal, const coordinateType belowVal, const coordinateType outsideVal)
	Thresholds scalar values of mesh in place. More...

static void	BinaryThresholdScalars (vtkSmartPointer< vtkPolyData > model, const coordinateType aboveVal, const coordinateType belowVal, const coordinateType insideVal=1.0, const coordinateType outsideVal=0.0)
	Thresholds scalar values of mesh in place. Scalars outside the given range is set to outsideVal and scalars inside the range set to insideVal. More...

static void	MaskScalars (vtkSmartPointer< vtkPolyData > model1, vtkSmartPointer< vtkPolyData > model2)
	Masks the scalars in model1 with scalars in model2. Only values corresponding to >= 1 in model2 are kept in model1.

static vtkSmartPointer< vtkMatrix4x4 >	ProcrustesAlign (vtkSmartPointer< vtkPolyData > source, vtkSmartPointer< vtkPolyData > target, bool rigid=false)
	Aligns the source mesh to target mesh assuming the meshes have the same number of points and correspondence. The transform is applied in-place to the source mesh. More...

Protected Member Functions
bool	IsCurrentModel ()

void	UpdateModelState (vtkSmartPointer< vtkPolyDataAlgorithm > filter)

void	UpdateModelStateFromModel (vtkSmartPointer< vtkPolyData > mesh)

Protected Attributes
vtkSmartPointer< vtkPolyData >	InputModel
	Holds the initial model in the pipeline.

vtkSmartPointer< vtkPolyData >	CurrentModel
	Holds the current model in the pipeline.

vtkSmartPointer< vtkPolyData >	PreviousModel
	Holds the previous model in the pipeline.

vtkSmartPointer< vtkTransform >	CurrentTransform
	transform applied to model

bool	InternalInPlaceOperation
	Used by the collection members to assign via pointers rather than deep copys.

Detailed Description

Represents a model (i.e. a model with cells and scalar values) and their common operations. Also allows batch operations on collection of models.

The class members are divided into two groups, atmoic and collection members. Atomic operations operate on one model at a time, whereas collection members use the atomic operations in a loop to batch that operation over multiple models.

Unlike the milx::Image class, this class tracks the current state of the data using pointers which can be retrieved using the Result() (or GetOutput()) and PreviousResult() members. This class is used extensively throughout the milxQtModel class and in milxSurfaceDiagnosticApp application. See the usage examples given below:

Processing Pipeline

milx::Model model;
        model.SetInput(surface);
        model.Triangulate();
        model.LaplacianSmoothing(250); //Kill some of the steps caused by MCs
        model.WindowedSincSmoothing(20); //Preserve features
        model.Clean();
        model.Update();

ICP Registration

...
vtkSmartPointer<vtkPolyData> mesh = m_RawSurfaces->GetNextItem();
milx::Model initialisation(m_AtlasSurface);
initialisation.MatchInformation(mesh, true); //Get correct approx scale
initialisation.IterativeClosestPointsRegistration(mesh, false); //Doesn't handle different FoV's

Landmark based registration

milx::Model model(m_AtlasSurface);
model.LandmarkBasedRegistration(mesh, true); 
mesh->DeepCopy(model.GetOutput());

Voxelise model

milx::Model model1(surface1);

vtkSmartPointer<vtkImageData> voxelisedModel1 = model1.Voxelise(255, labelSpacing.GetDataPointer(), bigBounds);

Multi-use example

//Read labels and generate iso surface
//Clip mesh to ensure same FoV
vtkSmartPointer<vtkFloatArray> weights = vtkSmartPointer<vtkFloatArray>::New();
milx::Model model(surface);
model.MarkSurfaceInsideImage<FloatImageType>(surface, distanceMap, weights); //sets weights as scalars in here
model.Clip(1.0, 1.0);
//Save distances as scalars on mesh
const bool absValues = true;
vtkSmartPointer<vtkFloatArray> scalars = model.SurfaceScalarsFromImage<FloatImageType>(model.GetOutput(), distanceMap, absValues);
scalars->SetName("Surface Distance Errors");
model.SetScalars(scalars);
//Restore correspondence to copy scalars over to full (unclipped) mesh, since the mesh is clipped to ensure same FoV
cout << "Regions of Surface Outside the Image are marked with -1." << endl;
weights->FillComponent(0, -1.0);
for(int j = 0; j < model.GetOutput()->GetNumberOfPoints(); j ++)
{
  vtkIdType index = surface->FindPoint(model.GetOutput()->GetPoint(j));
  weights->SetTuple1(index, scalars->GetTuple1(j));
}

Definition at line 113 of file milxModel.h.

The documentation for this class was generated from the following files:

include/milxModel.h
src/milxModel.cxx

Public Member Functions

Static Public Member Functions

Protected Member Functions

Protected Attributes

Detailed Description