Point Cloud Library (PCL)  1.14.1-dev
color_modality.h
1 /*
2  * Software License Agreement (BSD License)
3  *
4  * Point Cloud Library (PCL) - www.pointclouds.org
5  * Copyright (c) 2010-2011, Willow Garage, Inc.
6  *
7  * All rights reserved.
8  *
9  * Redistribution and use in source and binary forms, with or without
10  * modification, are permitted provided that the following conditions
11  * are met:
12  *
13  * * Redistributions of source code must retain the above copyright
14  * notice, this list of conditions and the following disclaimer.
15  * * Redistributions in binary form must reproduce the above
16  * copyright notice, this list of conditions and the following
17  * disclaimer in the documentation and/or other materials provided
18  * with the distribution.
19  * * Neither the name of Willow Garage, Inc. nor the names of its
20  * contributors may be used to endorse or promote products derived
21  * from this software without specific prior written permission.
22  *
23  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
24  * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
25  * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
26  * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
27  * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
28  * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
29  * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
30  * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
31  * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
32  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
33  * ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
34  * POSSIBILITY OF SUCH DAMAGE.
35  *
36  */
37 
38 #pragma once
39 
40 #include <pcl/recognition/quantizable_modality.h>
41 #include <pcl/recognition/distance_map.h>
42 
43 #include <pcl/pcl_base.h>
44 #include <pcl/point_cloud.h>
45 #include <pcl/point_types.h>
46 #include <pcl/recognition/point_types.h>
47 
48 #include <algorithm>
49 #include <cstddef>
50 #include <list>
51 #include <vector>
52 
53 namespace pcl
54 {
55 
56  // --------------------------------------------------------------------------
57 
58  template <typename PointInT>
60  : public QuantizableModality, public PCLBase<PointInT>
61  {
62  protected:
64 
65  struct Candidate
66  {
67  float distance;
68 
69  unsigned char bin_index;
70 
71  std::size_t x;
72  std::size_t y;
73 
74  bool
75  operator< (const Candidate & rhs)
76  {
77  return (distance > rhs.distance);
78  }
79  };
80 
81  public:
83 
84  ColorModality ();
85 
86  virtual ~ColorModality () = default;
87 
88  inline QuantizedMap &
90  {
91  return (filtered_quantized_colors_);
92  }
93 
94  inline QuantizedMap &
96  {
97  return (spreaded_filtered_quantized_colors_);
98  }
99 
100  void
101  extractFeatures (const MaskMap & mask, std::size_t nr_features, std::size_t modalityIndex,
102  std::vector<QuantizedMultiModFeature> & features) const;
103 
104  /** \brief Provide a pointer to the input dataset (overwrites the PCLBase::setInputCloud method)
105  * \param cloud the const boost shared pointer to a PointCloud message
106  */
107  virtual void
108  setInputCloud (const typename PointCloudIn::ConstPtr & cloud)
109  {
110  input_ = cloud;
111  }
112 
113  virtual void
114  processInputData ();
115 
116  protected:
117 
118  void
119  quantizeColors ();
120 
121  void
123 
124  static inline int
125  quantizeColorOnRGBExtrema (const float r,
126  const float g,
127  const float b);
128 
129  void
130  computeDistanceMap (const MaskMap & input, DistanceMap & output) const;
131 
132  private:
133  float feature_distance_threshold_;
134 
135  pcl::QuantizedMap quantized_colors_;
136  pcl::QuantizedMap filtered_quantized_colors_;
137  pcl::QuantizedMap spreaded_filtered_quantized_colors_;
138 
139  };
140 
141 }
142 
143 //////////////////////////////////////////////////////////////////////////////////////////////
144 template <typename PointInT>
146  : feature_distance_threshold_ (1.0f), quantized_colors_ (), filtered_quantized_colors_ (), spreaded_filtered_quantized_colors_ ()
147 {
148 }
149 
150 //////////////////////////////////////////////////////////////////////////////////////////////
151 template <typename PointInT>
152 void
154 {
155  // quantize gradients
156  quantizeColors ();
157 
158  // filter quantized gradients to get only dominants one + thresholding
159  filterQuantizedColors ();
160 
161  // spread filtered quantized gradients
162  //spreadFilteredQunatizedColorGradients ();
163  const int spreading_size = 8;
164  pcl::QuantizedMap::spreadQuantizedMap (filtered_quantized_colors_,
165  spreaded_filtered_quantized_colors_, spreading_size);
166 }
167 
168 //////////////////////////////////////////////////////////////////////////////////////////////
169 template <typename PointInT>
171  const std::size_t nr_features,
172  const std::size_t modality_index,
173  std::vector<QuantizedMultiModFeature> & features) const
174 {
175  const std::size_t width = mask.getWidth ();
176  const std::size_t height = mask.getHeight ();
177 
178  MaskMap mask_maps[8];
179  for (std::size_t map_index = 0; map_index < 8; ++map_index)
180  mask_maps[map_index].resize (width, height);
181 
182  unsigned char map[255]{};
183 
184  map[0x1<<0] = 0;
185  map[0x1<<1] = 1;
186  map[0x1<<2] = 2;
187  map[0x1<<3] = 3;
188  map[0x1<<4] = 4;
189  map[0x1<<5] = 5;
190  map[0x1<<6] = 6;
191  map[0x1<<7] = 7;
192 
193  QuantizedMap distance_map_indices (width, height);
194  //memset (distance_map_indices.data, 0, sizeof (distance_map_indices.data[0])*width*height);
195 
196  for (std::size_t row_index = 0; row_index < height; ++row_index)
197  {
198  for (std::size_t col_index = 0; col_index < width; ++col_index)
199  {
200  if (mask (col_index, row_index) != 0)
201  {
202  //const unsigned char quantized_value = quantized_surface_normals_ (row_index, col_index);
203  const unsigned char quantized_value = filtered_quantized_colors_ (col_index, row_index);
204 
205  if (quantized_value == 0)
206  continue;
207  const int dist_map_index = map[quantized_value];
208 
209  distance_map_indices (col_index, row_index) = dist_map_index;
210  //distance_maps[dist_map_index].at<unsigned char>(row_index, col_index) = 255;
211  mask_maps[dist_map_index] (col_index, row_index) = 255;
212  }
213  }
214  }
215 
216  DistanceMap distance_maps[8];
217  for (int map_index = 0; map_index < 8; ++map_index)
218  computeDistanceMap (mask_maps[map_index], distance_maps[map_index]);
219 
220  std::list<Candidate> list1;
221  std::list<Candidate> list2;
222 
223  float weights[8] = {0,0,0,0,0,0,0,0};
224 
225  const std::size_t off = 4;
226  for (std::size_t row_index = off; row_index < height-off; ++row_index)
227  {
228  for (std::size_t col_index = off; col_index < width-off; ++col_index)
229  {
230  if (mask (col_index, row_index) != 0)
231  {
232  //const unsigned char quantized_value = quantized_surface_normals_ (row_index, col_index);
233  const unsigned char quantized_value = filtered_quantized_colors_ (col_index, row_index);
234 
235  //const float nx = surface_normals_ (col_index, row_index).normal_x;
236  //const float ny = surface_normals_ (col_index, row_index).normal_y;
237  //const float nz = surface_normals_ (col_index, row_index).normal_z;
238 
239  if (quantized_value != 0)
240  {
241  const int distance_map_index = map[quantized_value];
242 
243  //const float distance = distance_maps[distance_map_index].at<float> (row_index, col_index);
244  const float distance = distance_maps[distance_map_index] (col_index, row_index);
245 
246  if (distance >= feature_distance_threshold_)
247  {
248  Candidate candidate;
249 
250  candidate.distance = distance;
251  candidate.x = col_index;
252  candidate.y = row_index;
253  candidate.bin_index = distance_map_index;
254 
255  list1.push_back (candidate);
256 
257  ++weights[distance_map_index];
258  }
259  }
260  }
261  }
262  }
263 
264  for (typename std::list<Candidate>::iterator iter = list1.begin (); iter != list1.end (); ++iter)
265  iter->distance *= 1.0f / weights[iter->bin_index];
266 
267  list1.sort ();
268 
269  if (list1.size () <= nr_features)
270  {
271  features.reserve (list1.size ());
272  for (typename std::list<Candidate>::iterator iter = list1.begin (); iter != list1.end (); ++iter)
273  {
274  QuantizedMultiModFeature feature;
275 
276  feature.x = static_cast<int> (iter->x);
277  feature.y = static_cast<int> (iter->y);
278  feature.modality_index = modality_index;
279  feature.quantized_value = filtered_quantized_colors_ (iter->x, iter->y);
280 
281  features.push_back (feature);
282  }
283 
284  return;
285  }
286 
287  int distance = static_cast<int> (list1.size () / nr_features + 1); // ??? @todo:!:!:!:!:!:!
288  while (list2.size () != nr_features)
289  {
290  const int sqr_distance = distance*distance;
291  for (typename std::list<Candidate>::iterator iter1 = list1.begin (); iter1 != list1.end (); ++iter1)
292  {
293  bool candidate_accepted = true;
294 
295  for (typename std::list<Candidate>::iterator iter2 = list2.begin (); iter2 != list2.end (); ++iter2)
296  {
297  const int dx = static_cast<int> (iter1->x) - static_cast<int> (iter2->x);
298  const int dy = static_cast<int> (iter1->y) - static_cast<int> (iter2->y);
299  const int tmp_distance = dx*dx + dy*dy;
300 
301  if (tmp_distance < sqr_distance)
302  {
303  candidate_accepted = false;
304  break;
305  }
306  }
307 
308  if (candidate_accepted)
309  list2.push_back (*iter1);
310 
311  if (list2.size () == nr_features) break;
312  }
313  --distance;
314  }
315 
316  for (typename std::list<Candidate>::iterator iter2 = list2.begin (); iter2 != list2.end (); ++iter2)
317  {
318  QuantizedMultiModFeature feature;
319 
320  feature.x = static_cast<int> (iter2->x);
321  feature.y = static_cast<int> (iter2->y);
322  feature.modality_index = modality_index;
323  feature.quantized_value = filtered_quantized_colors_ (iter2->x, iter2->y);
324 
325  features.push_back (feature);
326  }
327 }
328 
329 //////////////////////////////////////////////////////////////////////////////////////////////
330 template <typename PointInT>
331 void
333 {
334  const std::size_t width = input_->width;
335  const std::size_t height = input_->height;
336 
337  quantized_colors_.resize (width, height);
338 
339  for (std::size_t row_index = 0; row_index < height; ++row_index)
340  {
341  for (std::size_t col_index = 0; col_index < width; ++col_index)
342  {
343  const float r = static_cast<float> ((*input_) (col_index, row_index).r);
344  const float g = static_cast<float> ((*input_) (col_index, row_index).g);
345  const float b = static_cast<float> ((*input_) (col_index, row_index).b);
346 
347  quantized_colors_ (col_index, row_index) = quantizeColorOnRGBExtrema (r, g, b);
348  }
349  }
350 }
351 
352 //////////////////////////////////////////////////////////////////////////////////////////////
353 template <typename PointInT>
354 void
356 {
357  const std::size_t width = input_->width;
358  const std::size_t height = input_->height;
359 
360  filtered_quantized_colors_.resize (width, height);
361 
362  // filter data
363  for (std::size_t row_index = 1; row_index < height-1; ++row_index)
364  {
365  for (std::size_t col_index = 1; col_index < width-1; ++col_index)
366  {
367  unsigned char histogram[8] = {0,0,0,0,0,0,0,0};
368 
369  {
370  const unsigned char * data_ptr = quantized_colors_.getData () + (row_index-1)*width+col_index-1;
371  assert (0 <= data_ptr[0] && data_ptr[0] < 9 &&
372  0 <= data_ptr[1] && data_ptr[1] < 9 &&
373  0 <= data_ptr[2] && data_ptr[2] < 9);
374  ++histogram[data_ptr[0]];
375  ++histogram[data_ptr[1]];
376  ++histogram[data_ptr[2]];
377  }
378  {
379  const unsigned char * data_ptr = quantized_colors_.getData () + row_index*width+col_index-1;
380  assert (0 <= data_ptr[0] && data_ptr[0] < 9 &&
381  0 <= data_ptr[1] && data_ptr[1] < 9 &&
382  0 <= data_ptr[2] && data_ptr[2] < 9);
383  ++histogram[data_ptr[0]];
384  ++histogram[data_ptr[1]];
385  ++histogram[data_ptr[2]];
386  }
387  {
388  const unsigned char * data_ptr = quantized_colors_.getData () + (row_index+1)*width+col_index-1;
389  assert (0 <= data_ptr[0] && data_ptr[0] < 9 &&
390  0 <= data_ptr[1] && data_ptr[1] < 9 &&
391  0 <= data_ptr[2] && data_ptr[2] < 9);
392  ++histogram[data_ptr[0]];
393  ++histogram[data_ptr[1]];
394  ++histogram[data_ptr[2]];
395  }
396 
397  unsigned char max_hist_value = 0;
398  int max_hist_index = -1;
399 
400  // for (int i = 0; i < 8; ++i)
401  // {
402  // if (max_hist_value < histogram[i+1])
403  // {
404  // max_hist_index = i;
405  // max_hist_value = histogram[i+1]
406  // }
407  // }
408  // Unrolled for performance optimization:
409  if (max_hist_value < histogram[0]) {max_hist_index = 0; max_hist_value = histogram[0];}
410  if (max_hist_value < histogram[1]) {max_hist_index = 1; max_hist_value = histogram[1];}
411  if (max_hist_value < histogram[2]) {max_hist_index = 2; max_hist_value = histogram[2];}
412  if (max_hist_value < histogram[3]) {max_hist_index = 3; max_hist_value = histogram[3];}
413  if (max_hist_value < histogram[4]) {max_hist_index = 4; max_hist_value = histogram[4];}
414  if (max_hist_value < histogram[5]) {max_hist_index = 5; max_hist_value = histogram[5];}
415  if (max_hist_value < histogram[6]) {max_hist_index = 6; max_hist_value = histogram[6];}
416  if (max_hist_value < histogram[7]) {max_hist_index = 7; max_hist_value = histogram[7];}
417 
418  //if (max_hist_index != -1 && max_hist_value >= 5)
419  filtered_quantized_colors_ (col_index, row_index) = 0x1 << max_hist_index;
420  //else
421  // filtered_quantized_color_gradients_ (col_index, row_index) = 0;
422 
423  }
424  }
425 }
426 
427 //////////////////////////////////////////////////////////////////////////////////////////////
428 template <typename PointInT>
429 int
431  const float g,
432  const float b)
433 {
434  const float r_inv = 255.0f-r;
435  const float g_inv = 255.0f-g;
436  const float b_inv = 255.0f-b;
437 
438  const float dist_0 = (r*r + g*g + b*b)*2.0f;
439  const float dist_1 = r*r + g*g + b_inv*b_inv;
440  const float dist_2 = r*r + g_inv*g_inv+ b*b;
441  const float dist_3 = r*r + g_inv*g_inv + b_inv*b_inv;
442  const float dist_4 = r_inv*r_inv + g*g + b*b;
443  const float dist_5 = r_inv*r_inv + g*g + b_inv*b_inv;
444  const float dist_6 = r_inv*r_inv + g_inv*g_inv+ b*b;
445  const float dist_7 = (r_inv*r_inv + g_inv*g_inv + b_inv*b_inv)*1.5f;
446 
447  const float min_dist = std::min (std::min (std::min (dist_0, dist_1), std::min (dist_2, dist_3)), std::min (std::min (dist_4, dist_5), std::min (dist_6, dist_7)));
448 
449  if (min_dist == dist_0)
450  {
451  return 0;
452  }
453  if (min_dist == dist_1)
454  {
455  return 1;
456  }
457  if (min_dist == dist_2)
458  {
459  return 2;
460  }
461  if (min_dist == dist_3)
462  {
463  return 3;
464  }
465  if (min_dist == dist_4)
466  {
467  return 4;
468  }
469  if (min_dist == dist_5)
470  {
471  return 5;
472  }
473  if (min_dist == dist_6)
474  {
475  return 6;
476  }
477  return 7;
478 }
479 
480 //////////////////////////////////////////////////////////////////////////////////////////////
481 template <typename PointInT> void
483  DistanceMap & output) const
484 {
485  const std::size_t width = input.getWidth ();
486  const std::size_t height = input.getHeight ();
487 
488  output.resize (width, height);
489 
490  // compute distance map
491  //float *distance_map = new float[input_->size ()];
492  const unsigned char * mask_map = input.getData ();
493  float * distance_map = output.getData ();
494  for (std::size_t index = 0; index < width*height; ++index)
495  {
496  if (mask_map[index] == 0)
497  distance_map[index] = 0.0f;
498  else
499  distance_map[index] = static_cast<float> (width + height);
500  }
501 
502  // first pass
503  float * previous_row = distance_map;
504  float * current_row = previous_row + width;
505  for (std::size_t ri = 1; ri < height; ++ri)
506  {
507  for (std::size_t ci = 1; ci < width; ++ci)
508  {
509  const float up_left = previous_row [ci - 1] + 1.4f; //distance_map[(ri-1)*input_->width + ci-1] + 1.4f;
510  const float up = previous_row [ci] + 1.0f; //distance_map[(ri-1)*input_->width + ci] + 1.0f;
511  const float up_right = previous_row [ci + 1] + 1.4f; //distance_map[(ri-1)*input_->width + ci+1] + 1.4f;
512  const float left = current_row [ci - 1] + 1.0f; //distance_map[ri*input_->width + ci-1] + 1.0f;
513  const float center = current_row [ci]; //distance_map[ri*input_->width + ci];
514 
515  const float min_value = std::min (std::min (up_left, up), std::min (left, up_right));
516 
517  if (min_value < center)
518  current_row[ci] = min_value; //distance_map[ri * input_->width + ci] = min_value;
519  }
520  previous_row = current_row;
521  current_row += width;
522  }
523 
524  // second pass
525  float * next_row = distance_map + width * (height - 1);
526  current_row = next_row - width;
527  for (int ri = static_cast<int> (height)-2; ri >= 0; --ri)
528  {
529  for (int ci = static_cast<int> (width)-2; ci >= 0; --ci)
530  {
531  const float lower_left = next_row [ci - 1] + 1.4f; //distance_map[(ri+1)*input_->width + ci-1] + 1.4f;
532  const float lower = next_row [ci] + 1.0f; //distance_map[(ri+1)*input_->width + ci] + 1.0f;
533  const float lower_right = next_row [ci + 1] + 1.4f; //distance_map[(ri+1)*input_->width + ci+1] + 1.4f;
534  const float right = current_row [ci + 1] + 1.0f; //distance_map[ri*input_->width + ci+1] + 1.0f;
535  const float center = current_row [ci]; //distance_map[ri*input_->width + ci];
536 
537  const float min_value = std::min (std::min (lower_left, lower), std::min (right, lower_right));
538 
539  if (min_value < center)
540  current_row[ci] = min_value; //distance_map[ri*input_->width + ci] = min_value;
541  }
542  next_row = current_row;
543  current_row -= width;
544  }
545 }
void computeDistanceMap(const MaskMap &input, DistanceMap &output) const
virtual void setInputCloud(const typename PointCloudIn::ConstPtr &cloud)
Provide a pointer to the input dataset (overwrites the PCLBase::setInputCloud method)
QuantizedMap & getSpreadedQuantizedMap()
Returns a reference to the internally computed spread quantized map.
virtual ~ColorModality()=default
static int quantizeColorOnRGBExtrema(const float r, const float g, const float b)
QuantizedMap & getQuantizedMap()
Returns a reference to the internally computed quantized map.
virtual void processInputData()
void extractFeatures(const MaskMap &mask, std::size_t nr_features, std::size_t modalityIndex, std::vector< QuantizedMultiModFeature > &features) const
Extracts features from this modality within the specified mask.
Represents a distance map obtained from a distance transformation.
Definition: distance_map.h:47
float * getData()
Returns a pointer to the beginning of map.
Definition: distance_map.h:70
void resize(const std::size_t width, const std::size_t height)
Resizes the map to the specified size.
Definition: distance_map.h:80
std::size_t getWidth() const
Definition: mask_map.h:57
unsigned char * getData()
Definition: mask_map.h:69
std::size_t getHeight() const
Definition: mask_map.h:63
PCL base class.
Definition: pcl_base.h:70
PointCloudConstPtr input_
The input point cloud dataset.
Definition: pcl_base.h:147
shared_ptr< const PointCloud< PointInT > > ConstPtr
Definition: point_cloud.h:414
Interface for a quantizable modality.
static void spreadQuantizedMap(const QuantizedMap &input_map, QuantizedMap &output_map, std::size_t spreading_size)
Defines all the PCL implemented PointT point type structures.
float distance(const PointT &p1, const PointT &p2)
Definition: geometry.h:60
bool operator<(const Candidate &rhs)
Feature that defines a position and quantized value in a specific modality.
std::size_t modality_index
the index of the corresponding modality.
unsigned char quantized_value
the quantized value attached to the feature.