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
|
/*
* Copyright (C) 2005-2020 Centre National d'Etudes Spatiales (CNES)
* Copyright (C) 2007-2012 Institut Mines Telecom / Telecom Bretagne
*
* This file is part of Orfeo Toolbox
*
* https://www.orfeo-toolbox.org/
*
* 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
*
* 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 otbCzihoSOMLearningBehaviorFunctor_h
#define otbCzihoSOMLearningBehaviorFunctor_h
#include "itkSize.h"
#include "otbMath.h"
namespace otb
{
namespace Functor
{
/** \class CzihoSOMLearningBehaviorFunctor
* \brief Beta behavior over SOM training phase
*
* This class implements an evolution of the \f$ \beta \f$ weightening
* coefficient over the SOM training. It is issued from A. Cziho's PhD:
* "Compression d'images et analyse de contenu par quantification vectorielle"
* PhD dissertation, University of Rennes I, Rennes, France. May 5th, 1999.
*
* Its behavior is decomposed into two steps depending on the number of iterations:
* \f[
\beta =
\begin{cases}
\beta_0 \left( 1 - \frac{t}{t_0} \right) & \textrm{ if } t < t_0 \\
\beta_{\textrm{end}} \left( 1- \frac{t-t_O}{t_{\textrm{end}}-t_0} \right) & \textrm{ if } t_0 \leqslant t < t_{\textrm{end}}
\end{cases}
\f]
* where \f$ t_0 \f$ stands for IterationThreshold.
*
* CzihoSOMLearningBehaviorFunctor uses some parameters of the SOM class such as:
* BetaInit, BetaEnd, NumberOfIterations, but also NeighborhoodSizeInit which may be
* (surprisingly) required for the IterationThreshold.
*
* The functor function uses \code NumberOfIterations \endcode, \code BetaInit \endcode, \code BetaEnd \endcode parameters, that is
* why it is necessary to call a specific method for \code IterationThreshold \endcode initialization.
*
* \sa SOM
*
* \ingroup OTBSOM
*/
class CzihoSOMLearningBehaviorFunctor
{
public:
/** Empty constructor / descructor */
CzihoSOMLearningBehaviorFunctor()
{
m_IterationThreshold = 0;
}
virtual ~CzihoSOMLearningBehaviorFunctor()
{
}
/** Accessors */
unsigned int GetIterationThreshold()
{
return this->m_IterationThreshold;
}
template <unsigned int VDimension>
void SetIterationThreshold(const itk::Size<VDimension>& sizeInit, unsigned int iterMax)
{
double V0 = static_cast<double>(sizeInit[0]);
for (unsigned int i = 1; i < VDimension; ++i)
{
if (V0 < static_cast<double>(sizeInit[i]))
V0 = static_cast<double>(sizeInit[i]);
}
m_IterationThreshold = static_cast<unsigned int>(static_cast<double>(iterMax) * (1.0 - 1.0 / ::std::sqrt(V0)));
}
/** Functor */
virtual double operator()(unsigned int currentIteration, unsigned int numberOfIterations, double betaInit, double betaEnd) const
{
if (currentIteration < m_IterationThreshold)
{
return betaInit * (1.0 - static_cast<double>(currentIteration) / static_cast<double>(numberOfIterations));
}
else
{
return betaEnd * (1.0 - static_cast<double>(currentIteration - m_IterationThreshold) / static_cast<double>(numberOfIterations - m_IterationThreshold));
}
}
private:
unsigned int m_IterationThreshold;
}; // end of class CzihoSOMLearningBehaviorFunctor
} // end namespace Functor
} // end namespace otb
#endif
|
