File: SIFTExample.cxx

package info (click to toggle)
otb 6.6.1%2Bdfsg-1
  • links: PTS, VCS
  • area: main
  • in suites: buster
  • size: 47,068 kB
  • sloc: cpp: 316,755; ansic: 4,474; sh: 1,610; python: 497; perl: 92; makefile: 82; java: 72
file content (296 lines) | stat: -rw-r--r-- 10,402 bytes parent folder | download
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
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
/*
 * Copyright (C) 2005-2017 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.
 */


//  Software Guide : BeginCommandLineArgs
//    INPUTS: {ROISpot5.png}
//    OUTPUTS: {ROISpot5SIFT0.png}
//    "SIFT0.txt" 2 3 0 0
//  Software Guide : EndCommandLineArgs
//  Software Guide : BeginCommandLineArgs
//    INPUTS: {ROISpot5.png}
//    OUTPUTS: {ROISpot5SIFT1.png}
//    "SIFT1.txt" 2 3 1 0
//  Software Guide : EndCommandLineArgs
//  Software Guide : BeginCommandLineArgs
//    INPUTS: {ROISpot5.png}
//    OUTPUTS: {ROISpot5SIFT2.png}
//    "SIFT2.txt" 2 3 2 0
//  Software Guide : EndCommandLineArgs
//  Software Guide : BeginCommandLineArgs
//    INPUTS: {QB_Suburb.png}
//    OUTPUTS: {QB_SuburbSIFT5.png}
//    "SIFT2.txt" 2 3 5 0
//  Software Guide : EndCommandLineArgs
//  Software Guide : BeginCommandLineArgs
//    INPUTS: {QB_SuburbRotated10.png}
//    OUTPUTS: {QB_SuburbSIFT5Rotated10.png}
//    "SIFT2.txt" 2 3 5 0
//  Software Guide : EndCommandLineArgs

// Software Guide : BeginLatex
//
// This example illustrates the use of the \doxygen{otb}{ImageToSIFTKeyPointSetFilter}.
// The Scale-Invariant Feature Transform (or SIFT) is an algorithm in
// computer vision to detect and describe local features in
// images. The algorithm was published by David Lowe
// \cite{LoweSIFT}. The detection and description of local image
// features can help in object recognition and image registration. The
// SIFT features are local and based on the appearance of the object
// at particular interest points, and are invariant to image scale and
// rotation. They are also robust to changes in illumination, noise,
// occlusion and minor changes in viewpoint.
//
// The first step required to use this filter is to include its header file.
//
// Software Guide : EndLatex

// Software Guide : BeginCodeSnippet
#include "otbImageToSIFTKeyPointSetFilter.h"
// Software Guide : EndCodeSnippet
#include "otbImage.h"
#include "otbImageFileReader.h"
#include "otbImageFileWriter.h"
#include "itkPointSet.h"
#include "itkVariableLengthVector.h"
#include "itkRGBPixel.h"
#include "itkImageRegionIterator.h"

#include <iostream>
#include <fstream>

int main(int argc, char * argv[])
{
  if (argc != 8)
    {
    std::cerr << "Usage: " << argv[0];
    std::cerr <<
    " InputImage OutputImage OutputSIFTFile octaves scales threshold ratio" <<
    std::endl;
    return 1;
    }
  const char * infname = argv[1];
  const char * outfname = argv[3];
  const char * outputImageFilename = argv[2];

  const unsigned int octaves = atoi(argv[4]);
  const unsigned int scales = atoi(argv[5]);
  float              threshold = atof(argv[6]);
  float              ratio = atof(argv[7]);

  typedef float RealType;
  const unsigned int Dimension = 2;

// Software Guide : BeginLatex
// The \doxygen{otb}{ImageToSIFTKeyPointSetFilter} is templated over
// its input image type and the output point set type. Therefore, we
// start by defining the needed types.
// Software Guide : EndLatex
// Software Guide : BeginCodeSnippet
  typedef otb::Image<RealType, Dimension>     ImageType;
  typedef itk::VariableLengthVector<RealType> RealVectorType;
  typedef otb::ImageFileReader<ImageType>     ReaderType;
  typedef itk::PointSet<RealVectorType,
      Dimension>            PointSetType;

  typedef otb::ImageToSIFTKeyPointSetFilter<ImageType,
      PointSetType>
  ImageToSIFTKeyPointSetFilterType;
// Software Guide : EndCodeSnippet

// Software Guide : BeginLatex
// Since the SIFT detector produces a point set, we will need
// iterators for the coordinates of the points and the data associated
// with them.
// Software Guide : EndLatex
// Software Guide : BeginCodeSnippet
  typedef PointSetType::PointsContainer    PointsContainerType;
  typedef PointsContainerType::Iterator    PointsIteratorType;
// Software Guide : EndCodeSnippet

// Software Guide : BeginLatex
// We can now instantiate the reader and the SIFT filter and plug the pipeline.
// Software Guide : EndLatex
// Software Guide : BeginCodeSnippet
  ReaderType::Pointer                       reader = ReaderType::New();
  ImageToSIFTKeyPointSetFilterType::Pointer filter =
    ImageToSIFTKeyPointSetFilterType::New();

  reader->SetFileName(infname);

  filter->SetInput(reader->GetOutput());
// Software Guide : EndCodeSnippet

// Software Guide : BeginLatex
// The SIFT filter needs the following parameters:
// \begin{itemize}
// \item the number of octaves, that is, the number of levels of undersampling,
// \item the number of scales (blurring) per octave,
// \item the low contrast threshold to be applied to each point for the detection
// on the difference of Gaussians image,
// \item the threshold on the responses to consider a point as an edge.
// \end{itemize}
// Software Guide : EndLatex
// Software Guide : BeginCodeSnippet
  filter->SetOctavesNumber(octaves);
  filter->SetScalesNumber(scales);

  filter->SetDoGThreshold(threshold);
  filter->SetEdgeThreshold(ratio);
// Software Guide : EndCodeSnippet

  filter->Update();

  //  Software Guide : BeginLatex
  // Figure~\ref{fig:SIFT} shows the result of applying the SIFT
  // point detector to a small patch extracted from a Spot 5 image
  // using different threshold values.
  // \begin{figure}
  // \center
  // \includegraphics[width=0.22\textwidth]{ROISpot5.eps}
  // \includegraphics[width=0.22\textwidth]{ROISpot5SIFT0.eps}
  // \includegraphics[width=0.22\textwidth]{ROISpot5SIFT1.eps}
  // \includegraphics[width=0.22\textwidth]{ROISpot5SIFT2.eps}
  // \itkcaption[SIFT Application]{Result of applying the
  // \doxygen{otb}{ImageToSIFTKeyPointSetFilter} to a Spot 5
  // image. Left to right: original image and SIFT with thresholds 0,
  // 1 and 2 respectively.}
  // \label{fig:SIFT}
  // \end{figure}
  // Figure~\ref{fig:SIFT2} shows the result of applying the SIFT
  // point detector to a small patch extracted from a Spot 5 image
  // using different threshold values.
  // \begin{figure}
  // \center
  // \includegraphics[width=0.30\textwidth]{QB_Suburb.eps}
  // \includegraphics[width=0.30\textwidth]{QB_SuburbSIFT5.eps}
  // \includegraphics[width=0.30\textwidth]{QB_SuburbSIFT5Rotated10.eps}
  // \itkcaption[SIFT Application]{Result of applying the
  // \doxygen{otb}{ImageToSIFTKeyPointSetFilter} to a high resolution image
  // image. Left to right: original image and SIFT on the original
  // and a rotated image respectively.}
  // \label{fig:SIFT2}
  // \end{figure}
  //  Software Guide : EndLatex

  //
  //Building the output image for visualization
  ImageType::OffsetType t = {{ 0, 1}};
  ImageType::OffsetType b = {{ 0, -1}};
  ImageType::OffsetType r = {{ 1, 0}};
  ImageType::OffsetType l = {{-1, 0}};

  typedef itk::RGBPixel<unsigned char> RGBPixelType;
  typedef otb::Image<RGBPixelType, 2>  OutputImageType;

  typedef otb::ImageFileWriter<OutputImageType> WriterType;

  OutputImageType::Pointer outputImage = OutputImageType::New();

  OutputImageType::RegionType region;
  OutputImageType::SizeType   outputSize;
  outputSize = reader->GetOutput()->GetLargestPossibleRegion().GetSize();
  region.SetSize(outputSize);

  OutputImageType::IndexType indexStart;
  indexStart[0] = 0;
  indexStart[1] = 0;
  region.SetIndex(indexStart);

  outputImage->SetRegions(region);
  outputImage->Allocate();

  itk::ImageRegionIterator<OutputImageType> iterOutput(outputImage,
                                                       reader->GetOutput()->
                                                       GetLargestPossibleRegion());

  for (iterOutput.GoToBegin(); !iterOutput.IsAtEnd(); ++iterOutput)
    {
    ImageType::IndexType       index = iterOutput.GetIndex();
    ImageType::PixelType       grayPix = reader->GetOutput()->GetPixel(index);
    OutputImageType::PixelType rgbPixel;
    rgbPixel.SetRed(static_cast<unsigned char>(grayPix));
    rgbPixel.SetGreen(static_cast<unsigned char>(grayPix));
    rgbPixel.SetBlue(static_cast<unsigned char>(grayPix));

    iterOutput.Set(rgbPixel);
    }

  PointsIteratorType        pIt = filter->GetOutput()->GetPoints()->Begin();
  ImageType::SpacingType    spacing = reader->GetOutput()->GetSignedSpacing();
  ImageType::PointType      origin = reader->GetOutput()->GetOrigin();
  OutputImageType::SizeType size =
    outputImage->GetLargestPossibleRegion().GetSize();

  while (pIt != filter->GetOutput()->GetPoints()->End())
    {
    ImageType::IndexType index;

    index[0] = (unsigned int)
               (vcl_floor
                  ((double) ((pIt.Value()[0] - origin[0]) / spacing[0] + 0.5)));

    index[1] = (unsigned int)
               (vcl_floor
                  ((double) ((pIt.Value()[1] - origin[1]) / spacing[1] + 0.5)));

    OutputImageType::PixelType keyPixel;
    keyPixel.SetRed(0);
    keyPixel.SetGreen(255);
    keyPixel.SetBlue(0);

    if (
      static_cast<unsigned int>(index[1]) <
      static_cast<unsigned int>(size[1])
      && static_cast<unsigned int>(index[0]) <
      static_cast<unsigned int>(size[0]))
      {
      outputImage->SetPixel(index, keyPixel);

      if (static_cast<unsigned int>(index[1]) <
          static_cast<unsigned int>(size[1] - 1))
        outputImage->SetPixel(
          index + t,
          keyPixel);

      if (index[1] > 0) outputImage->SetPixel(index + b, keyPixel);

      if (static_cast<unsigned int>(index[0]) <
          static_cast<unsigned int>(size[0] - 1))
        outputImage->SetPixel(
          index + r,
          keyPixel);

      if (index[0] > 0) outputImage->SetPixel(index + l, keyPixel);
      }
    ++pIt;
    }

  std::ofstream outfile(outfname);
  outfile << filter;
  outfile.close();

  WriterType::Pointer writer = WriterType::New();
  writer->SetInput(outputImage);
  writer->SetFileName(outputImageFilename);
  writer->Update();

  return EXIT_SUCCESS;
}