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
|
/*=========================================================================
*
* Copyright NumFOCUS
*
* 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
*
* https://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 itkCurvesLevelSetFunction_h
#define itkCurvesLevelSetFunction_h
#include "itkSegmentationLevelSetFunction.h"
namespace itk
{
/** \class CurvesLevelSetFunction
*
* \brief This function is used in CurvesLevelSetImageFilter to
* segment structures in images based on user supplied edge potential map.
*
* \par CurvesLevelSetFunction is a subclass of the generic LevelSetFunction.
* It is useful for segmentations based on a user supplied edge potential map which
* has values close to zero in regions near edges (or high image gradient) and values
* close to one in regions with relatively constant intensity. Typically, the edge
* potential map is a function of the gradient, for example:
*
* \f[ g(I) = 1 / ( 1 + | (\nabla * G)(I)| ) \f]
* \f[ g(I) = \exp^{-|(\nabla * G)(I)|} \f]
*
* where \f$ I \f$ is image intensity and
* \f$ (\nabla * G) \f$ is the derivative of Gaussian operator.
*
* \par In this function both the propagation term \f$ P(\mathbf{x}) \f$
* and the curvature spatial modifier term $\f$ Z(\mathbf{x}) \f$ are taken directly
* from the edge potential image. The edge potential image is set via the
* SetFeatureImage() method. An advection term \f$ A(\mathbf{x}) \f$ is constructed
* from the negative gradient of the edge potential image. This term behaves like
* a doublet attracting the contour to the edges.
*
* \par This implementation is based on:
* L. Lorigo, O. Faugeras, W.E.L. Grimson, R. Keriven, R. Kikinis, A. Nabavi,
* and C.-F. Westin, Curves: Curve evolution for vessel segmentation.
* Medical Image Analysis, 5:195-206, 2001.
*
* \sa LevelSetFunction
* \sa SegmentationLevelSetImageFunction
* \sa GeodesicActiveContourLevelSetImageFilter
*
* \ingroup FiniteDifferenceFunctions
* \ingroup ITKLevelSets
*/
template <typename TImageType, typename TFeatureImageType = TImageType>
class ITK_TEMPLATE_EXPORT CurvesLevelSetFunction : public SegmentationLevelSetFunction<TImageType, TFeatureImageType>
{
public:
ITK_DISALLOW_COPY_AND_MOVE(CurvesLevelSetFunction);
/** Standard class type aliases. */
using Self = CurvesLevelSetFunction;
using Superclass = SegmentationLevelSetFunction<TImageType, TFeatureImageType>;
using SuperSuperclass = LevelSetFunction<TImageType>;
using Pointer = SmartPointer<Self>;
using ConstPointer = SmartPointer<const Self>;
using FeatureImageType = TFeatureImageType;
/** Method for creation through the object factory. */
itkNewMacro(Self);
/** \see LightObject::GetNameOfClass() */
itkOverrideGetNameOfClassMacro(CurvesLevelSetFunction);
/** Extract some parameters from the superclass. */
using PixelType = typename SuperSuperclass::PixelType;
using typename Superclass::ImageType;
using typename Superclass::NeighborhoodType;
using typename Superclass::ScalarValueType;
using typename Superclass::FeatureScalarType;
using typename Superclass::RadiusType;
using FloatOffsetType = typename SuperSuperclass::FloatOffsetType;
using GlobalDataStruct = typename SuperSuperclass::GlobalDataStruct;
using typename Superclass::VectorImageType;
/** Extract some parameters from the superclass. */
static constexpr unsigned int ImageDimension = Superclass::ImageDimension;
/** Compute speed image from feature image. */
void
CalculateSpeedImage() override;
/** Compute the advection field from feature image. */
void
CalculateAdvectionImage() override;
/** The curvature speed is same as the propagation speed. */
ScalarValueType
CurvatureSpeed(const NeighborhoodType & neighborhood,
const FloatOffsetType & offset,
GlobalDataStruct * gd) const override
{
return this->PropagationSpeed(neighborhood, offset, gd);
}
/** Set/Get the sigma for the Gaussian kernel used to compute the gradient
* of the feature image needed for the advection term of the equation. */
void
SetDerivativeSigma(const double v)
{
m_DerivativeSigma = v;
}
double
GetDerivativeSigma()
{
return m_DerivativeSigma;
}
void
Initialize(const RadiusType & r) override;
protected:
CurvesLevelSetFunction()
{
// Curvature term is the minimal curvature.
this->UseMinimalCurvatureOn();
this->SetAdvectionWeight(NumericTraits<ScalarValueType>::OneValue());
this->SetPropagationWeight(NumericTraits<ScalarValueType>::OneValue());
this->SetCurvatureWeight(NumericTraits<ScalarValueType>::OneValue());
}
~CurvesLevelSetFunction() override = default;
void
PrintSelf(std::ostream & os, Indent indent) const override
{
Superclass::PrintSelf(os, indent);
os << indent << "DerivativeSigma: " << m_DerivativeSigma << std::endl;
}
private:
/** Slices for the ND neighborhood. */
std::slice x_slice[ImageDimension];
/** The offset of the center pixel in the neighborhood. */
OffsetValueType m_Center{ 0 };
/** Stride length along the y-dimension. */
OffsetValueType m_xStride[ImageDimension]{};
double m_DerivativeSigma{ 1.0 };
};
} // end namespace itk
#ifndef ITK_MANUAL_INSTANTIATION
# include "itkCurvesLevelSetFunction.hxx"
#endif
#endif
|
