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
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
|
/*=========================================================================
Program: Insight Segmentation & Registration Toolkit
Module: itkSampleClassifierFilterTest2.cxx
Language: C++
Date: $Date$
Version: $Revision$
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.
=========================================================================*/
#if defined(_MSC_VER)
#pragma warning ( disable : 4786 )
#endif
#include "itkArray.h"
#include "itkVariableLengthVector.h"
#include "itkListSample.h"
#include "itkSampleClassifierFilter.h"
#include "itkMaximumDecisionRule2.h"
#include "itkGaussianMembershipFunction.h"
#include "itkNormalVariateGenerator.h"
//Test if the SampleClassifier filter labels observations correctly
int itkSampleClassifierFilterTest2( int, char * [] )
{
const unsigned int numberOfComponents = 1;
typedef float MeasurementType;
const unsigned int numberOfClasses = 2;
typedef itk::Array< MeasurementType > MeasurementVectorType;
typedef itk::Statistics::ListSample< MeasurementVectorType > SampleType;
typedef itk::Statistics::SampleClassifierFilter< SampleType > FilterType;
FilterType::Pointer filter = FilterType::New();
SampleType::Pointer sample = SampleType::New();
sample->SetMeasurementVectorSize( numberOfComponents );
filter->SetNumberOfClasses( numberOfClasses );
if( filter->GetNumberOfClasses() != numberOfClasses )
{
std::cerr << "GetNumberOfClasses() didn't matched SetNumberOfClasses()" << std::endl;
return EXIT_FAILURE;
}
typedef FilterType::ClassLabelVectorObjectType ClassLabelVectorObjectType;
typedef FilterType::ClassLabelVectorType ClassLabelVectorType;
typedef FilterType::MembershipFunctionVectorObjectType MembershipFunctionVectorObjectType;
typedef FilterType::MembershipFunctionVectorType MembershipFunctionVectorType;
typedef itk::Statistics::GaussianMembershipFunction< MeasurementVectorType >
MembershipFunctionType;
typedef MembershipFunctionType::MeanType MeanType;
typedef MembershipFunctionType::CovarianceType CovarianceType;
typedef MembershipFunctionType::Pointer MembershipFunctionPointer;
ClassLabelVectorObjectType::Pointer classLabelsObject = ClassLabelVectorObjectType::New();
filter->SetClassLabels( classLabelsObject );
MembershipFunctionVectorObjectType::Pointer membershipFunctionsObject =
MembershipFunctionVectorObjectType::New();
filter->SetMembershipFunctions( membershipFunctionsObject );
// Add three membership functions and rerun the filter
MembershipFunctionVectorType & membershipFunctionsVector = membershipFunctionsObject->Get();
MembershipFunctionPointer membershipFunction1 = MembershipFunctionType::New();
membershipFunction1->SetMeasurementVectorSize( numberOfComponents );
MeanType mean1;
itk::Statistics::MeasurementVectorTraits::SetLength( mean1, numberOfComponents );
mean1[0] = 10.5;
membershipFunction1->SetMean( mean1 );
CovarianceType covariance1;
covariance1.SetSize( numberOfComponents, numberOfComponents );
covariance1.SetIdentity();
covariance1[0][0] = 0.5;
membershipFunction1->SetCovariance( covariance1 );
membershipFunctionsVector.push_back( membershipFunction1.GetPointer() );
MembershipFunctionPointer membershipFunction2 = MembershipFunctionType::New();
membershipFunction1->SetMeasurementVectorSize( numberOfComponents );
MeanType mean2;
itk::Statistics::MeasurementVectorTraits::SetLength( mean2, numberOfComponents );
mean2[0] = 200.5;
membershipFunction2->SetMean( mean2 );
CovarianceType covariance2;
covariance2.SetSize( numberOfComponents, numberOfComponents );
covariance2.SetIdentity();
covariance2[0][0] = 0.5;
membershipFunction2->SetCovariance( covariance2 );
membershipFunctionsVector.push_back( membershipFunction2.GetPointer() );
// Add class labels
ClassLabelVectorType & classLabelVector = classLabelsObject->Get();
typedef FilterType::ClassLabelType ClassLabelType;
ClassLabelType class1 = 0;
classLabelVector.push_back( class1 );
ClassLabelType class2 = 1;
classLabelVector.push_back( class2 );
//Set a decision rule type
typedef itk::Statistics::MaximumDecisionRule2 DecisionRuleType;
DecisionRuleType::Pointer decisionRule = DecisionRuleType::New();
filter->SetDecisionRule( decisionRule );
//Generate samples from a Gaussian distribution with mean and
//covariance parameter as the first membership function.
//All the samples should be labeled by the classifier as
//the first class
typedef itk::Statistics::NormalVariateGenerator NormalGeneratorType;
NormalGeneratorType::Pointer normalGenerator = NormalGeneratorType::New();
normalGenerator->Initialize( 101 );
MeasurementVectorType mv;
itk::Statistics::MeasurementVectorTraits::SetLength( mv, numberOfComponents );
double mean = mean1[0];
double standardDeviation = vcl_sqrt(covariance1[0][0]);
unsigned int numberOfSampleEachClass = 10;
for ( unsigned int i = 0; i < numberOfSampleEachClass; ++i )
{
mv[0] = (normalGenerator->GetVariate() * standardDeviation ) + mean;
sample->PushBack( mv );
}
//Add samples for the second gaussian
mean = mean2[0];
standardDeviation = vcl_sqrt(covariance1[0][0]);
for ( unsigned int i = 0; i < numberOfSampleEachClass; ++i )
{
mv[0] = (normalGenerator->GetVariate() * standardDeviation ) + mean;
sample->PushBack( mv );
}
filter->SetInput( sample );
try
{
filter->Update();
}
catch( itk::ExceptionObject & excp )
{
std::cerr << excp << std::endl;
return EXIT_FAILURE;
}
//Check if the measurement vectors are correctly labelled.
const FilterType::MembershipSampleType* membershipSample = filter->GetOutput();
FilterType::MembershipSampleType::ConstIterator iter = membershipSample->Begin();
unsigned int sampleCounter = 0;
while ( iter != membershipSample->End() )
{
if( sampleCounter < numberOfSampleEachClass )
{
if( iter.GetClassLabel() != class1 )
{
std::cerr << "Classification error: " << sampleCounter
<< "\t" << iter.GetClassLabel()
<< "\tclass1=" << class1 << std::endl;
return EXIT_FAILURE;
}
}
else
{
if( iter.GetClassLabel() != class2 )
{
std::cerr << "Classification error: " << sampleCounter
<< "\t" << iter.GetClassLabel()
<< "\tclass2=" << class2 << std::endl;
return EXIT_FAILURE;
}
}
++iter;
++sampleCounter;
}
std::cout << "Test passed." << std::endl;
return EXIT_SUCCESS;
}
|
