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
|
/*
* Copyright (C) 2005-2022 Centre National d'Etudes Spatiales (CNES)
*
* 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.
*/
#include "otbImageClassificationFilter.h"
#include "otbVectorImage.h"
#include "otbImage.h"
#include "otbImageFileReader.h"
#include "otbImageFileWriter.h"
#include "otbSharkRandomForestsMachineLearningModelFactory.h"
#include <random>
#include <chrono>
const unsigned int Dimension = 2;
typedef float PixelType;
typedef unsigned short LabeledPixelType;
typedef otb::VectorImage<PixelType, Dimension> ImageType;
typedef otb::Image<LabeledPixelType, Dimension> LabeledImageType;
typedef otb::ImageClassificationFilter<ImageType, LabeledImageType> ClassificationFilterType;
typedef ClassificationFilterType::ModelType ModelType;
typedef ClassificationFilterType::ValueType ValueType;
typedef ClassificationFilterType::LabelType LabelType;
typedef otb::SharkRandomForestsMachineLearningModelFactory<ValueType, LabelType> MachineLearningModelFactoryType;
typedef otb::ImageFileReader<ImageType> ReaderType;
typedef otb::ImageFileReader<LabeledImageType> MaskReaderType;
typedef otb::ImageFileWriter<LabeledImageType> WriterType;
typedef otb::SharkRandomForestsMachineLearningModel<PixelType, short unsigned int> MachineLearningModelType;
typedef MachineLearningModelType::InputValueType LocalInputValueType;
typedef MachineLearningModelType::InputSampleType LocalInputSampleType;
typedef MachineLearningModelType::InputListSampleType LocalInputListSampleType;
typedef MachineLearningModelType::TargetValueType LocalTargetValueType;
typedef MachineLearningModelType::TargetSampleType LocalTargetSampleType;
typedef MachineLearningModelType::TargetListSampleType LocalTargetListSampleType;
void generateSamples(unsigned int num_classes, unsigned int num_samples, unsigned int num_features, LocalInputListSampleType* samples,
LocalTargetListSampleType* labels)
{
std::default_random_engine generator;
std::uniform_int_distribution<int> label_distribution(1, num_classes);
std::uniform_int_distribution<int> feat_distribution(0, 256);
for (size_t scount = 0; scount < num_samples; ++scount)
{
LabeledPixelType label = label_distribution(generator);
LocalInputSampleType sample(num_features);
for (unsigned int i = 0; i < num_features; ++i)
sample[i] = feat_distribution(generator);
samples->SetMeasurementVectorSize(num_features);
samples->PushBack(sample);
labels->PushBack(label);
}
}
void buildModel(unsigned int num_classes, unsigned int num_samples, unsigned int num_features, std::string modelfname)
{
LocalInputListSampleType::Pointer samples = LocalInputListSampleType::New();
LocalTargetListSampleType::Pointer labels = LocalTargetListSampleType::New();
std::cout << "Sample generation\n";
generateSamples(num_classes, num_samples, num_features, samples, labels);
MachineLearningModelType::Pointer classifier = MachineLearningModelType::New();
classifier->SetInputListSample(samples);
classifier->SetTargetListSample(labels);
classifier->SetRegressionMode(false);
classifier->SetNumberOfTrees(100);
classifier->SetMTry(0);
classifier->SetNodeSize(25);
classifier->SetOobRatio(0.3);
std::cout << "Training\n";
using TimeT = std::chrono::milliseconds;
auto start = std::chrono::system_clock::now();
classifier->Train();
auto duration = std::chrono::duration_cast<TimeT>(std::chrono::system_clock::now() - start);
auto elapsed = duration.count();
std::cout << "Training took " << elapsed << " ms\n";
classifier->Save(modelfname);
}
int otbSharkImageClassificationFilter(int argc, char* argv[])
{
if (argc < 6 || argc > 8)
{
std::cout << "Usage: input_image output_image output_confidence output_proba batchmode [in_model_name] [mask_name]\n";
}
std::string imfname = argv[1];
std::string outfname = argv[2];
std::string conffname = argv[3];
std::string probafname = argv[4];
bool batch = (std::string(argv[5]) == "1");
std::string modelfname = "/tmp/rf_model.txt";
std::string maskfname{};
int num_classes = 3;
MaskReaderType::Pointer mask_reader = MaskReaderType::New();
ReaderType::Pointer reader = ReaderType::New();
reader->SetFileName(imfname);
reader->UpdateOutputInformation();
auto num_features = reader->GetOutput()->GetNumberOfComponentsPerPixel();
std::cout << "Image has " << num_features << " bands\n";
if (argc > 6)
{
modelfname = argv[6];
// We don't know the number of classes, so we set it to a high number
num_classes = 10;
}
else
{
buildModel(num_classes, 1000, num_features, modelfname);
}
ClassificationFilterType::Pointer filter = ClassificationFilterType::New();
MachineLearningModelType::Pointer model = MachineLearningModelType::New();
model->Load(modelfname);
filter->SetModel(model);
filter->SetInput(reader->GetOutput());
filter->SetNumberOfClasses(num_classes);
if (argc == 8)
{
maskfname = argv[7];
mask_reader->SetFileName(maskfname);
filter->SetInputMask(mask_reader->GetOutput());
}
WriterType::Pointer writer = WriterType::New();
writer->SetInput(filter->GetOutput());
writer->SetFileName(outfname);
std::cout << "Classification\n";
filter->SetBatchMode(batch);
filter->SetUseConfidenceMap(true);
filter->SetUseProbaMap(true);
using TimeT = std::chrono::milliseconds;
auto start = std::chrono::system_clock::now();
writer->Update();
auto duration = std::chrono::duration_cast<TimeT>(std::chrono::system_clock::now() - start);
auto elapsed = duration.count();
std::cout << "Classification took " << elapsed << " ms\n";
auto confWriter = otb::ImageFileWriter<ClassificationFilterType::ConfidenceImageType>::New();
confWriter->SetInput(filter->GetOutputConfidence());
confWriter->SetFileName(conffname);
confWriter->Update();
auto probaWriter = otb::ImageFileWriter<ClassificationFilterType::ProbaImageType>::New();
probaWriter->SetInput(filter->GetOutputProba());
probaWriter->SetFileName(probafname);
probaWriter->Update();
// Check that the chosen labels correspond to the max proba
itk::ImageRegionConstIterator<LabeledImageType> labIt(filter->GetOutput(), filter->GetOutput()->GetLargestPossibleRegion());
itk::ImageRegionConstIterator<ClassificationFilterType::ProbaImageType> probIt(filter->GetOutputProba(),
filter->GetOutputProba()->GetLargestPossibleRegion());
for (labIt.GoToBegin(), probIt.GoToBegin(); !labIt.IsAtEnd(); ++labIt, ++probIt)
{
if (labIt.Get() > 0) // Pixel is not masked
{
auto first = probIt.Get().GetDataPointer();
auto last = probIt.Get().GetDataPointer();
std::advance(last, num_classes);
auto max_proba = std::distance(first, std::max_element(first, last)) + 1;
if (labIt.Get() != max_proba)
{
std::cout << "Chosen label " << labIt.Get() << " and max proba position " << max_proba << " from " << probIt.Get() << " don't match\n";
return EXIT_FAILURE;
}
}
}
return EXIT_SUCCESS;
}
|
