1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244
|
/*=========================================================================
*
* Copyright Insight Software Consortium
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0.txt
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*
*=========================================================================*/
#ifndef itkMultiResolutionPyramidImageFilter_h
#define itkMultiResolutionPyramidImageFilter_h
#include "itkImageToImageFilter.h"
#include "itkArray2D.h"
namespace itk
{
/** \class MultiResolutionPyramidImageFilter
* \brief Framework for creating images in a multi-resolution
* pyramid.
*
* MultiResolutionPyramidImageFilter provides a generic framework to
* to create a image pryamid according to a user defined
* multi-resolution schedule.
*
* The multi-resolution schedule is specified in terms for
* shrink factors at each multi-resolution level for each dimension.
*
* A user can either use the default schedules or specify
* each factor in the schedules directly.
*
* The schedule is stored as an unsigned int matrix.
* An element of the table can be access via the double bracket
* notation: table[resLevel][dimension]
*
* For example:
* 8 4 4
* 4 4 2
*
* is a schedule for two computation level. In the first (coarest)
* level the image is reduce by a factor of 8 in the column dimension,
* factor of 4 in the row dimension and factor of 4 in the slice dimension.
* In the second level, the image is reduce by a factor of 4 in the column
* dimension, 4 is the row dimension and 2 in the slice dimension.
*
* The method SetNumberOfLevels() set the number of
* computation levels in the pyramid. This method will
* allocate memory for the multi-resolution schedule table.
* This method generates defaults tables with the starting
* shrink factor for all dimension set to 2^(NumberOfLevel - 1).
* All factors are halved for all subsequent levels.
* For example if the number of levels was set to 4, the default table is:
*
* 8 8 8
* 4 4 4
* 2 2 2
* 1 1 1
*
* The user can get a copy of the schedule via GetSchedule()
* They may make alteration and reset it using SetSchedule().
*
* A user can create a default table by specifying the starting
* shrink factors via methods SetStartingShrinkFactors()
* The factors for subsequent level is generated by
* halving the factor or setting to one, depending on which is larger.
*
* For example, for 4 levels and starting factors of 8,8,4
* the default table would be:
*
* 8 8 4
* 4 4 2
* 2 2 1
* 1 1 1
*
* When this filter is updated, NumberOfLevels outputs are produced.
* The N'th output correspond to the N'th level of the pyramid.
*
* To generate each output image, Gaussian smoothing is first performed
* using a DiscreteGaussianImageFilter with variance (shrink factor / 2)^2.
* The smoothed image is then downsampled using a ResampleImageFilter.
*
* Note that even if the shrink factors are all equal to one, a smoothing
* will still be applied. The output at the finest level of the pyramid
* will thus typically be a smoothed version of the input.
*
* This class is templated over the input image type and the output image
* type.
*
* This filter uses multithreaded filters to perform the smoothing and
* downsampling.
*
* This filter supports streaming.
*
* \sa DiscreteGaussianImageFilter
* \sa ShrinkImageFilter
*
* \ingroup PyramidImageFilter MultiThreaded Streamed
* \ingroup ITKRegistrationCommon
*/
template<
typename TInputImage,
typename TOutputImage
>
class ITK_TEMPLATE_EXPORT MultiResolutionPyramidImageFilter:
public ImageToImageFilter< TInputImage, TOutputImage >
{
public:
/** Standard class typedefs. */
typedef MultiResolutionPyramidImageFilter Self;
typedef ImageToImageFilter< TInputImage, TOutputImage > Superclass;
typedef SmartPointer< Self > Pointer;
typedef SmartPointer< const Self > ConstPointer;
/** Method for creation through the object factory. */
itkNewMacro(Self);
/** Run-time type information (and related methods). */
itkTypeMacro(MultiResolutionPyramidImageFilter, ImageToImageFilter);
/** ScheduleType typedef support. */
typedef Array2D< unsigned int > ScheduleType;
/** ImageDimension enumeration. */
itkStaticConstMacro(ImageDimension, unsigned int,
TInputImage::ImageDimension);
itkStaticConstMacro(OutputImageDimension, unsigned int,
TOutputImage::ImageDimension);
/** Inherit types from Superclass. */
typedef typename Superclass::InputImageType InputImageType;
typedef typename Superclass::OutputImageType OutputImageType;
typedef typename Superclass::InputImagePointer InputImagePointer;
typedef typename Superclass::OutputImagePointer OutputImagePointer;
typedef typename Superclass::InputImageConstPointer InputImageConstPointer;
/** Set the number of multi-resolution levels. The matrix containing the
* schedule will be resized accordingly. The schedule is populated with
* default values. At the coarsest (0) level, the shrink factors are set
* 2^(nlevel - 1) for all dimension. These shrink factors are halved for
* subsequent levels. The number of levels is clamped to a minimum value
* of 1. All shrink factors are also clamped to a minimum value of 1. */
virtual void SetNumberOfLevels(unsigned int num);
/** Get the number of multi-resolution levels. */
itkGetConstMacro(NumberOfLevels, unsigned int);
/** Set a multi-resolution schedule. The input schedule must have only
* ImageDimension number of columns and NumberOfLevels number of rows. For
* each dimension, the shrink factor must be non-increasing with respect to
* subsequent levels. This function will clamp shrink factors to satisify
* this condition. All shrink factors less than one will also be clamped
* to the value of 1. */
virtual void SetSchedule(const ScheduleType & schedule);
/** Get the multi-resolution schedule. */
itkGetConstReferenceMacro(Schedule, ScheduleType);
/** Set the starting shrink factor for the coarsest (0) resolution
* level. The schedule is then populated with defaults values obtained by
* halving the factors at the previous level. All shrink factors are
* clamped to a minimum value of 1. */
virtual void SetStartingShrinkFactors(unsigned int factor);
virtual void SetStartingShrinkFactors(unsigned int *factors);
/** Get the starting shrink factors */
const unsigned int * GetStartingShrinkFactors() const;
/** Test if the schedule is downward divisible. This method returns true if
* at every level, the shrink factors are divisble by the shrink factors at
* the next level. */
static bool IsScheduleDownwardDivisible(const ScheduleType & schedule);
/** MultiResolutionPyramidImageFilter produces images which are of
* different resolution and different pixel spacing than its input image.
* As such, MultiResolutionPyramidImageFilter needs to provide an
* implementation for GenerateOutputInformation() in order to inform the
* pipeline execution model. The original documentation of this method is
* below. \sa ProcessObject::GenerateOutputInformaton() */
virtual void GenerateOutputInformation() ITK_OVERRIDE;
/** Given one output whose requested region has been set, this method sets
* the requested region for the remaining output images. The original
* documentation of this method is below. \sa
* ProcessObject::GenerateOutputRequestedRegion(); */
virtual void GenerateOutputRequestedRegion(DataObject *output) ITK_OVERRIDE;
/** MultiResolutionPyramidImageFilter requires a larger input requested
* region than the output requested regions to accommodate the shrinkage and
* smoothing operations. As such, MultiResolutionPyramidImageFilter needs
* to provide an implementation for GenerateInputRequestedRegion(). The
* original documentation of this method is below. \sa
* ProcessObject::GenerateInputRequestedRegion() */
virtual void GenerateInputRequestedRegion() ITK_OVERRIDE;
itkSetMacro(MaximumError, double);
itkGetConstReferenceMacro(MaximumError, double);
itkSetMacro(UseShrinkImageFilter, bool);
itkGetConstMacro(UseShrinkImageFilter, bool);
itkBooleanMacro(UseShrinkImageFilter);
#ifdef ITK_USE_CONCEPT_CHECKING
// Begin concept checking
itkConceptMacro( SameDimensionCheck,
( Concept::SameDimension< ImageDimension, OutputImageDimension > ) );
itkConceptMacro( OutputHasNumericTraitsCheck,
( Concept::HasNumericTraits< typename TOutputImage::PixelType > ) );
// End concept checking
#endif
protected:
MultiResolutionPyramidImageFilter();
~MultiResolutionPyramidImageFilter() ITK_OVERRIDE {}
void PrintSelf(std::ostream & os, Indent indent) const ITK_OVERRIDE;
/** Generate the output data. */
void GenerateData() ITK_OVERRIDE;
double m_MaximumError;
unsigned int m_NumberOfLevels;
ScheduleType m_Schedule;
bool m_UseShrinkImageFilter;
private:
ITK_DISALLOW_COPY_AND_ASSIGN(MultiResolutionPyramidImageFilter);
};
} // namespace itk
#ifndef ITK_MANUAL_INSTANTIATION
#include "itkMultiResolutionPyramidImageFilter.hxx"
#endif
#endif
|