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
|
/*=========================================================================
Program: Insight Segmentation & Registration Toolkit
Module: $RCSfile: itkGaussianDerivativeSpatialFunction.txx,v $
Language: C++
Date: $Date: 2007-06-11 15:12:09 $
Version: $Revision: 1.9 $
Copyright (c) Insight Software Consortium. All rights reserved.
See ITKCopyright.txt or http://www.itk.org/HTML/Copyright.htm for details.
This software is distributed WITHOUT ANY WARRANTY; without even
the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR
PURPOSE. See the above copyright notices for more information.
=========================================================================*/
#ifndef __itkGaussianDerivativeSpatialFunction_txx
#define __itkGaussianDerivativeSpatialFunction_txx
#include <math.h>
#include "itkGaussianDerivativeSpatialFunction.h"
namespace itk
{
template <typename TOutput, unsigned int VImageDimension, typename TInput>
GaussianDerivativeSpatialFunction<TOutput, VImageDimension, TInput>
::GaussianDerivativeSpatialFunction()
{
m_Mean = ArrayType::Filled(0.0);
m_Sigma = ArrayType::Filled(1.0);
m_Scale = 1.0;
m_Normalized = false;
m_Direction = 0;
}
template <typename TOutput, unsigned int VImageDimension, typename TInput>
GaussianDerivativeSpatialFunction<TOutput, VImageDimension, TInput>
::~GaussianDerivativeSpatialFunction()
{
}
template <typename TOutput, unsigned int VImageDimension, typename TInput>
typename GaussianDerivativeSpatialFunction<TOutput, VImageDimension, TInput>::OutputType
GaussianDerivativeSpatialFunction<TOutput, VImageDimension, TInput>
::Evaluate(const TInput& position) const
{
// Normalizing the Gaussian is important for statistical applications
// but is generally not desirable for creating images because of the
// very small numbers involved (would need to use doubles)
double prefixDenom;
if (m_Normalized)
{
prefixDenom = m_Sigma[m_Direction]*m_Sigma[m_Direction];
for(unsigned int i = 0; i < VImageDimension; i++)
{
prefixDenom *= m_Sigma[i];
}
prefixDenom *= 2*vcl_pow( 2 * 3.1415927, VImageDimension / 2.0);
}
else
{
prefixDenom = 1.0;
}
double suffixExp = 0;
for(unsigned int i = 0; i < VImageDimension; i++)
{
suffixExp += (position[m_Direction] - m_Mean[m_Direction])*(position[m_Direction] - m_Mean[m_Direction]) / (2 * m_Sigma[m_Direction] * m_Sigma[m_Direction]);
}
double value = -2*(position[m_Direction] - m_Mean[m_Direction])*m_Scale * (1 / prefixDenom) * vcl_exp(-1 * suffixExp);
return (TOutput) value;
}
/** Evaluate the function at a given position and return a vector */
template <typename TOutput, unsigned int VImageDimension, typename TInput>
typename GaussianDerivativeSpatialFunction<TOutput, VImageDimension, TInput>::VectorType
GaussianDerivativeSpatialFunction<TOutput, VImageDimension, TInput>
::EvaluateVector(const TInput& position) const
{
VectorType gradient;
for(unsigned int i = 0; i < VImageDimension; i++)
{
m_Direction = i;
gradient[i]=this->Evaluate(position);
}
return gradient;
}
template <typename TOutput, unsigned int VImageDimension, typename TInput>
void
GaussianDerivativeSpatialFunction<TOutput, VImageDimension, TInput>
::PrintSelf(std::ostream& os, Indent indent) const
{
Superclass::PrintSelf(os,indent);
os << indent << "Sigma: " << m_Sigma << std::endl;
os << indent << "Mean: " << m_Mean << std::endl;
os << indent << "Scale: " << m_Scale << std::endl;
os << indent << "Normalized?: " << m_Normalized << std::endl;
os << indent << "Direction: " << m_Direction << std::endl;
}
} // end namespace itk
#endif
|
