template<typename TFixedImage, typename TMovingImage, typename TDisplacementField, typename TParentImageFilter = PDEDeformableRegistrationFilter<TFixedImage, TMovingImage, TDisplacementField>>
class itk::GPUPDEDeformableRegistrationFilter< TFixedImage, TMovingImage, TDisplacementField, TParentImageFilter >
Deformably register two images using a PDE algorithm.
GPUPDEDeformableRegistrationFilter is a base case for filter implementing a PDE deformable algorithm that register two images by computing the deformation field which will map a moving image onto a fixed image.
A deformation field is represented as a image whose pixel type is some vector type with at least N elements, where N is the dimension of the fixed image. The vector type must support element access via operator []. It is assumed that the vector elements behave like floating point scalars.
This class is templated over the fixed image type, moving image type and the deformation Field type.
The input fixed and moving images are set via methods SetFixedImage and SetMovingImage respectively. An initial deformation field maybe set via SetInitialDisplacementField or SetInput. If no initial field is set, a zero field is used as the initial condition.
The output deformation field can be obtained via methods GetOutput or GetDisplacementField.
The PDE algorithm is run for a user defined number of iterations. Typically the PDE algorithm requires period Gaussian smoothing of the deformation field to enforce an elastic-like condition. The amount of smoothing is governed by a set of user defined standard deviations (one for each dimension).
In terms of memory, this filter keeps two internal buffers: one for storing the intermediate updates to the field and one for double-buffering when smoothing the deformation field. Both buffers are the same type and size as the output deformation field.
This class make use of the finite difference solver hierarchy. Update for each iteration is computed using a PDEDeformableRegistrationFunction.
- Warning
- This filter assumes that the fixed image type, moving image type and deformation field type all have the same number of dimensions.
- See also
- PDEDeformableRegistrationFunction.
Definition at line 82 of file itkGPUPDEDeformableRegistrationFilter.h.
|
| using | ConstPointer = SmartPointer< const Self > |
| |
| using | CPUSuperclass = TParentImageFilter |
| |
| using | DeformationScalarType = typename TDisplacementField::PixelType::ValueType |
| |
| using | DeformationVectorType = typename TDisplacementField::PixelType |
| |
| using | DisplacementFieldPointer = typename DisplacementFieldType::Pointer |
| |
| using | DisplacementFieldType = TDisplacementField |
| |
| using | FiniteDifferenceFunctionType = typename GPUSuperclass::FiniteDifferenceFunctionType |
| |
| using | FixedImageConstPointer = typename FixedImageType::ConstPointer |
| |
| using | FixedImagePointer = typename FixedImageType::Pointer |
| |
| using | FixedImageType = TFixedImage |
| |
| using | GPUPDEDeformableRegistrationFunctionType = GPUPDEDeformableRegistrationFunction< FixedImageType, MovingImageType, DisplacementFieldType > |
| |
| using | GPUSuperclass = GPUDenseFiniteDifferenceImageFilter< TDisplacementField, TDisplacementField, TParentImageFilter > |
| |
| using | MovingImageConstPointer = typename MovingImageType::ConstPointer |
| |
| using | MovingImagePointer = typename MovingImageType::Pointer |
| |
| using | MovingImageType = TMovingImage |
| |
| using | OutputImageType = typename GPUSuperclass::OutputImageType |
| |
| using | Pointer = SmartPointer< Self > |
| |
| using | Self = GPUPDEDeformableRegistrationFilter |
| |
| using | StandardDeviationsType = FixedArray< double, ImageDimension > |
| |
| using | ConstPointer = SmartPointer< const Self > |
| |
| using | CPUSuperclass = TParentImageFilter |
| |
| using | FiniteDifferenceFunctionType = typename GPUSuperclass::FiniteDifferenceFunctionType |
| |
| using | GPUSuperclass = GPUFiniteDifferenceImageFilter< TDisplacementField, TDisplacementField, TParentImageFilter > |
| |
| using | InputImageType = typename GPUSuperclass::InputImageType |
| |
| using | OutputImageType = typename GPUSuperclass::OutputImageType |
| |
| using | PixelType = typename GPUSuperclass::PixelType |
| |
| using | Pointer = SmartPointer< Self > |
| |
| using | Self = GPUDenseFiniteDifferenceImageFilter |
| |
| using | TimeStepType = typename GPUSuperclass::TimeStepType |
| |
| using | UpdateBufferType = OutputImageType |
| |
| using | ConstPointer = SmartPointer< const Self > |
| |
| using | CPUSuperclass = TParentImageFilter |
| |
| using | FiniteDifferenceFunctionType = typename GPUFiniteDifferenceFunction< TDisplacementField >::DifferenceFunctionType |
| |
| using | GPUSuperclass = GPUInPlaceImageFilter< TDisplacementField, TDisplacementField, TParentImageFilter > |
| |
| using | InputImageType = TDisplacementField |
| |
| using | InputPixelType = typename TDisplacementField ::PixelType |
| |
| using | InputPixelValueType = typename NumericTraits< InputPixelType >::ValueType |
| |
| using | NeighborhoodScalesType = typename FiniteDifferenceFunctionType::NeighborhoodScalesType |
| |
| using | OutputImageType = TDisplacementField |
| |
| using | OutputPixelType = typename TDisplacementField ::PixelType |
| |
| using | OutputPixelValueType = typename NumericTraits< OutputPixelType >::ValueType |
| |
| using | PixelType = OutputPixelType |
| |
| using | Pointer = SmartPointer< Self > |
| |
| using | RadiusType = typename FiniteDifferenceFunctionType::RadiusType |
| |
| using | Self = GPUFiniteDifferenceImageFilter |
| |
| using | TimeStepType = typename FiniteDifferenceFunctionType::TimeStepType |
| |
| using | ConstPointer = SmartPointer< const Self > |
| |
| using | CPUSuperclass = TParentImageFilter |
| |
| using | GPUSuperclass = GPUImageToImageFilter< TDisplacementField, TDisplacementField, TParentImageFilter > |
| |
| using | InputImageConstPointer = typename InputImageType::ConstPointer |
| |
| using | InputImagePixelType = typename InputImageType::PixelType |
| |
| using | InputImagePointer = typename InputImageType::Pointer |
| |
| using | InputImageRegionType = typename InputImageType::RegionType |
| |
| using | InputImageType = TDisplacementField |
| |
| using | OutputImagePixelType = typename GPUSuperclass::OutputImagePixelType |
| |
| using | OutputImagePointer = typename GPUSuperclass::OutputImagePointer |
| |
| using | OutputImageRegionType = typename GPUSuperclass::OutputImageRegionType |
| |
| using | OutputImageType = typename GPUSuperclass::OutputImageType |
| |
| using | Pointer = SmartPointer< Self > |
| |
| using | Self = GPUInPlaceImageFilter |
| |
| using | ConstPointer = SmartPointer< const Self > |
| |
| using | InputImageConstPointer = typename InputImageType::ConstPointer |
| |
| using | InputImagePixelType = typename InputImageType::PixelType |
| |
| using | InputImagePointer = typename InputImageType::Pointer |
| |
| using | InputImageRegionType = typename InputImageType::RegionType |
| |
| using | InputImageType = TDisplacementField |
| |
| using | Pointer = SmartPointer< Self > |
| |
| using | Self = GPUImageToImageFilter |
| |
| using | Superclass = TParentImageFilter |
| |
Inheritance diagram for itk::GPUPDEDeformableRegistrationFilter< TFixedImage, TMovingImage, TDisplacementField, TParentImageFilter >:
1.8.16