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
|
#ifndef GMMCLASSIFYIMAGEFILTER_TXX
#define GMMCLASSIFYIMAGEFILTER_TXX
#include "GMMClassifyImageFilter.h"
#include "itkImageRegionConstIterator.h"
#include "EMGaussianMixtures.h"
#include "ImageCollectionToImageFilter.h"
template <class TInputImage, class TInputVectorImage, class TOutputImage>
GMMClassifyImageFilter<TInputImage, TInputVectorImage, TOutputImage>
::GMMClassifyImageFilter()
{
m_MixtureModel = NULL;
}
template <class TInputImage, class TInputVectorImage, class TOutputImage>
GMMClassifyImageFilter<TInputImage, TInputVectorImage, TOutputImage>
::~GMMClassifyImageFilter()
{
}
template <class TInputImage, class TInputVectorImage, class TOutputImage>
void
GMMClassifyImageFilter<TInputImage, TInputVectorImage, TOutputImage>
::SetMixtureModel(GaussianMixtureModel *model)
{
m_MixtureModel = model;
this->Modified();
}
template <class TInputImage, class TInputVectorImage, class TOutputImage>
void
GMMClassifyImageFilter<TInputImage, TInputVectorImage, TOutputImage>
::AddScalarImage(InputImageType *image)
{
this->AddInput(image);
}
template <class TInputImage, class TInputVectorImage, class TOutputImage>
void
GMMClassifyImageFilter<TInputImage, TInputVectorImage, TOutputImage>
::AddVectorImage(InputVectorImageType *image)
{
this->AddInput(image);
}
template <class TInputImage, class TInputVectorImage, class TOutputImage>
void
GMMClassifyImageFilter<TInputImage, TInputVectorImage, TOutputImage>
::GenerateInputRequestedRegion()
{
itk::ImageSource<TOutputImage>::GenerateInputRequestedRegion();
for( itk::InputDataObjectIterator it( this ); !it.IsAtEnd(); it++ )
{
// Check whether the input is an image of the appropriate dimension
InputImageType *input = dynamic_cast< InputImageType * >( it.GetInput() );
InputVectorImageType *vecInput = dynamic_cast< InputVectorImageType * >( it.GetInput() );
if (input)
{
InputImageRegionType inputRegion;
this->CallCopyOutputRegionToInputRegion( inputRegion, this->GetOutput()->GetRequestedRegion() );
input->SetRequestedRegion(inputRegion);
}
else if(vecInput)
{
InputImageRegionType inputRegion;
this->CallCopyOutputRegionToInputRegion( inputRegion, this->GetOutput()->GetRequestedRegion() );
vecInput->SetRequestedRegion(inputRegion);
}
}
}
template <class TInputImage, class TInputVectorImage, class TOutputImage>
void
GMMClassifyImageFilter<TInputImage, TInputVectorImage, TOutputImage>
::PrintSelf(std::ostream &os, itk::Indent indent) const
{
os << indent << "GMMClassifyImageFilter" << std::endl;
}
template <class TInputImage, class TInputVectorImage, class TOutputImage>
void
GMMClassifyImageFilter<TInputImage, TInputVectorImage, TOutputImage>
::ThreadedGenerateData(const OutputImageRegionType &outputRegionForThread,
itk::ThreadIdType threadId)
{
// Get the number of inputs
assert(m_MixtureModel);
int n_input = this->GetNumberOfIndexedInputs();
OutputImagePointer outputPtr = this->GetOutput(0);
// Create a collection iterator
typedef ImageCollectionConstRegionIteratorWithIndex<
TInputImage, TInputVectorImage> CollectionIter;
typedef itk::ImageRegionIterator<TOutputImage> OutputIter;
OutputIter it_out(outputPtr, outputRegionForThread);
vnl_vector<double> x(m_MixtureModel->GetNumberOfComponents());
vnl_vector<double> x_scratch(m_MixtureModel->GetNumberOfComponents());
vnl_vector<double> z_scratch(m_MixtureModel->GetNumberOfComponents());
vnl_vector<double> log_pdf(m_MixtureModel->GetNumberOfGaussians());
vnl_vector<double> log_w(m_MixtureModel->GetNumberOfGaussians());
vnl_vector<double> w(m_MixtureModel->GetNumberOfGaussians());
vnl_vector<double> p(m_MixtureModel->GetNumberOfGaussians());
// Create a multiplier vector (1 for foreground, -1 for background)
vnl_vector<double> pfactor(m_MixtureModel->GetNumberOfGaussians());
for(int i = 0; i < m_MixtureModel->GetNumberOfGaussians(); i++)
{
pfactor[i] = m_MixtureModel->IsForeground(i) ? 1.0 : -1.0;
log_w[i] = log(m_MixtureModel->GetWeight(i));
w[i] = m_MixtureModel->GetWeight(i);
}
// Configure the input collection iterator
CollectionIter cit(outputRegionForThread);
for( itk::InputDataObjectIterator it( this ); !it.IsAtEnd(); it++ )
cit.AddImage(it.GetInput());
// Get the number of components
int nComp = cit.GetTotalComponents();
// Iterate through all the voxels
while ( !it_out.IsAtEnd() )
{
for(int i = 0; i < nComp; i++)
{
x[i] = cit.Value(i);
}
// Evaluate the posterior probability robustly
for(int k = 0; k < m_MixtureModel->GetNumberOfGaussians(); k++)
{
log_pdf[k] = m_MixtureModel->EvaluateLogPDF(k, x, x_scratch);
}
// Evaluate the GMM for each of the clusters
double pdiff = 0;
for(int k = 0; k < m_MixtureModel->GetNumberOfGaussians(); k++)
{
p[k] = EMGaussianMixtures::ComputePosterior(
m_MixtureModel->GetNumberOfGaussians(),
log_pdf.data_block(), w.data_block(), log_w.data_block(), k);
pdiff += p[k] * pfactor[k];
}
// Store the value
it_out.Set((OutputPixelType)(pdiff * 0x7fff));
++it_out;
++cit;
}
}
#endif
|
