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
|
/*=========================================================================
*
* Copyright UMC Utrecht and contributors
*
* 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 elxOptimizerBase_h
#define elxOptimizerBase_h
/** Needed for the macros */
#include "elxMacro.h"
#include "elxBaseComponentSE.h"
#include "itkOptimizer.h"
namespace elastix
{
/**
* \class OptimizerBase
* \brief This class is the elastix base class for all Optimizers.
*
* This class contains all the common functionality for Optimizers.
*
* The parameters used in this class are:
* \parameter NewSamplesEveryIteration: if this flag is set to "true" some
* optimizers ask the metric to select a new set of spatial samples in
* every iteration. This, if used in combination with the correct optimizer (such as the
* StandardGradientDescent), and ImageSampler (Random, RandomCoordinate, or RandomSparseMask),
* allows for a very low number of spatial samples (around 2000), even with large images
* and transforms with a large number of parameters.\n
* Choose one from {"true", "false"} for every resolution.\n
* example: <tt>(NewSamplesEveryIteration "true" "true" "true")</tt> \n
* Default is "false" for every resolution.\n
*
* \ingroup Optimizers
* \ingroup ComponentBaseClasses
*/
template <class TElastix>
class ITK_TEMPLATE_EXPORT OptimizerBase : public BaseComponentSE<TElastix>
{
public:
ITK_DISALLOW_COPY_AND_MOVE(OptimizerBase);
/** Standard ITK-stuff. */
using Self = OptimizerBase;
using Superclass = BaseComponentSE<TElastix>;
/** Run-time type information (and related methods). */
itkTypeMacro(OptimizerBase, BaseComponentSE);
/** Typedefs inherited from Elastix. */
using typename Superclass::ElastixType;
using typename Superclass::RegistrationType;
/** ITKBaseType. */
using ITKBaseType = itk::Optimizer;
/** Typedef needed for the SetCurrentPositionPublic function. */
using ParametersType = typename ITKBaseType::ParametersType;
/** Retrieves this object as ITKBaseType. */
ITKBaseType *
GetAsITKBaseType()
{
return &(this->GetSelf());
}
/** Retrieves this object as ITKBaseType, to use in const functions. */
const ITKBaseType *
GetAsITKBaseType() const
{
return &(this->GetSelf());
}
/** Add empty SetCurrentPositionPublic, so this function is known in every inherited class. */
virtual void
SetCurrentPositionPublic(const ParametersType & param);
/** Execute stuff before each new pyramid resolution:
* \li Find out if new samples are used every new iteration in this resolution.
*/
void
BeforeEachResolutionBase() override;
/** Execute stuff after registration:
* \li Compute and print MD5 hash of the transform parameters.
*/
void
AfterRegistrationBase() override;
/** Method that sets the scales defined by a sinus
* scale[i] = amplitude^( sin(i/nrofparam*2pi*frequency) )
*/
virtual void
SetSinusScales(double amplitude, double frequency, unsigned long numberOfParameters);
protected:
/** The constructor. */
OptimizerBase() = default;
/** The destructor. */
~OptimizerBase() override = default;
/** Force the metric to base its computation on a new subset of image samples.
* Not every metric may have implemented this.
*/
virtual void
SelectNewSamples();
/** Check whether the user asked to select new samples every iteration. */
virtual bool
GetNewSamplesEveryIteration() const;
struct SettingsType
{
double a, A, alpha, fmax, fmin, omega;
};
using SettingsVectorType = typename std::vector<SettingsType>;
/** Print the contents of the settings vector to log::info. */
static void
PrintSettingsVector(const SettingsVectorType & settings);
private:
elxDeclarePureVirtualGetSelfMacro(ITKBaseType);
/** Member variable to store the user preference for using new
* samples each iteration.
*/
bool m_NewSamplesEveryIteration{ false };
};
} // end namespace elastix
#ifndef ITK_MANUAL_INSTANTIATION
# include "elxOptimizerBase.hxx"
#endif
#endif // end #ifndef elxOptimizerBase_h
|
