C++ PointCloudOut::clear方法代码示例

template <typename PointInT, typename PointNT, typename PointOutT> void
pcl::GFPFHEstimation<PointInT, PointNT, PointOutT>::computeFeature (PointCloudOut &output)
{
    pcl::octree::OctreePointCloudSearch<PointInT> octree (octree_leaf_size_);
    octree.setInputCloud (input_);
    octree.addPointsFromInputCloud ();

    typename pcl::PointCloud<PointInT>::VectorType occupied_cells;
    octree.getOccupiedVoxelCenters (occupied_cells);

    // Determine the voxels crosses along the line segments
    // formed by every pair of occupied cells.
    std::vector< std::vector<int> > line_histograms;
    for (size_t i = 0; i < occupied_cells.size (); ++i)
    {
        Eigen::Vector3f origin = occupied_cells[i].getVector3fMap ();

        for (size_t j = i+1; j < occupied_cells.size (); ++j)
        {
            typename pcl::PointCloud<PointInT>::VectorType intersected_cells;
            Eigen::Vector3f end = occupied_cells[j].getVector3fMap ();
            octree.getApproxIntersectedVoxelCentersBySegment (origin, end, intersected_cells, 0.5f);

            // Intersected cells are ordered from closest to furthest w.r.t. the origin.
            std::vector<int> histogram;
            for (size_t k = 0; k < intersected_cells.size (); ++k)
            {
                std::vector<int> indices;
                octree.voxelSearch (intersected_cells[k], indices);
                int label = emptyLabel ();
                if (indices.size () != 0)
                {
                    label = getDominantLabel (indices);
                }
                histogram.push_back (label);
            }

            line_histograms.push_back(histogram);
        }
    }

    std::vector< std::vector<int> > transition_histograms;
    computeTransitionHistograms (line_histograms, transition_histograms);

    std::vector<float> distances;
    computeDistancesToMean (transition_histograms, distances);

    std::vector<float> gfpfh_histogram;
    computeDistanceHistogram (distances, gfpfh_histogram);

    output.clear ();
    output.width = 1;
    output.height = 1;
    output.points.resize (1);
    std::copy (gfpfh_histogram.begin (), gfpfh_histogram.end (), output.points[0].histogram);
}

//.........这里部分代码省略.........
        std::vector<float> r (4,0);

        r[0] = squaredNormalsDiff (up, center);
        r[0]+= squaredNormalsDiff (down, center);

        r[1] = squaredNormalsDiff (upright, center);
        r[1]+= squaredNormalsDiff (downleft, center);

        r[2] = squaredNormalsDiff (right, center);
        r[2]+= squaredNormalsDiff (left, center);

        r[3] = squaredNormalsDiff (downright, center);
        r[3]+= squaredNormalsDiff (upleft, center);

        float d = *(std::min_element (r.begin (), r.end ()));

        if (d < first_threshold_) continue;

        std::vector<float> B (4,0);
        std::vector<float> A (4,0);
        std::vector<float> sumAB (4,0);
        B[0] = normalsDiff (upright, up) * normalsDiff (up, center);
        B[0]+= normalsDiff (downleft, down) * normalsDiff (down, center);
        B[1] = normalsDiff (right, upright) * normalsDiff (upright, center);
        B[1]+= normalsDiff (left, downleft) * normalsDiff (downleft, center);
        B[2] = normalsDiff (downright, right) * normalsDiff (downright, center);
        B[2]+= normalsDiff (upleft, left) * normalsDiff (upleft, center);
        B[3] = normalsDiff (down, downright) * normalsDiff (downright, center);
        B[3]+= normalsDiff (up, upleft) * normalsDiff (upleft, center);
        A[0] = r[1] - r[0] - B[0] - B[0];
        A[1] = r[2] - r[1] - B[1] - B[1];
        A[2] = r[3] - r[2] - B[2] - B[2];
        A[3] = r[0] - r[3] - B[3] - B[3];
        sumAB[0] = A[0] + B[0];
        sumAB[1] = A[1] + B[1];
        sumAB[2] = A[2] + B[2];
        sumAB[3] = A[3] + B[3];
        if ((*std::max_element (B.begin (), B.end ()) < 0) &&
            (*std::min_element (sumAB.begin (), sumAB.end ()) > 0))
        {
          std::vector<float> D (4,0);
          D[0] = B[0] * B[0] / A[0];
          D[1] = B[1] * B[1] / A[1];
          D[2] = B[2] * B[2] / A[2];
          D[3] = B[3] * B[3] / A[3];
          response (i,j) = *(std::min (D.begin (), D.end ()));
        }
        else
          response (i,j) = d;
      }
    }
  }
  // Non maximas suppression
  std::vector<int> indices = *indices_;
  std::sort (indices.begin (), indices.end (),
             boost::bind (&TrajkovicKeypoint3D::greaterCornernessAtIndices, this, _1, _2));

  output.clear ();
  output.reserve (input_->size ());

  std::vector<bool> occupency_map (indices.size (), false);
  const int width (input_->width);
  const int height (input_->height);
  const int occupency_map_size (indices.size ());

#ifdef _OPENMP
#pragma omp parallel for shared (output) num_threads (threads_)
#endif
  for (int i = 0; i < indices.size (); ++i)
  {
    int idx = indices[i];
    if ((response_->points[idx] < second_threshold_) || occupency_map[idx])
      continue;

    PointOutT p;
    p.getVector3fMap () = input_->points[idx].getVector3fMap ();
    p.intensity = response_->points [idx];

#ifdef _OPENMP
#pragma omp critical
#endif
    {
      output.push_back (p);
      keypoints_indices_->indices.push_back (idx);
    }

    const int x = idx % width;
    const int y = idx / width;
    const int u_end = std::min (width, x + half_window_size_);
    const int v_end = std::min (height, y + half_window_size_);
    for(int v = std::max (0, y - half_window_size_); v < v_end; ++v)
      for(int u = std::max (0, x - half_window_size_); u < u_end; ++u)
        occupency_map[v*width + u] = true;
  }

  output.height = 1;
  output.width = static_cast<uint32_t> (output.size());
  // we don not change the denseness
  output.is_dense = true;
}

template <typename PointInT, typename PointOutT, typename IntensityT> void
pcl::HarrisKeypoint2D<PointInT, PointOutT, IntensityT>::detectKeypoints (PointCloudOut &output)
{
  derivatives_cols_.resize (input_->width, input_->height);
  derivatives_rows_.resize (input_->width, input_->height);
  //Compute cloud intensities first derivatives along columns and rows
  //!!! nsallem 20120220 : we don't test here for density so if one term in nan the result is nan
  int w = static_cast<int> (input_->width) - 1;
  int h = static_cast<int> (input_->height) - 1;
  // j = 0 --> j-1 out of range ; use 0
  // i = 0 --> i-1 out of range ; use 0
  derivatives_cols_(0,0) = (intensity_ ((*input_) (0,1)) - intensity_ ((*input_) (0,0))) * 0.5;
  derivatives_rows_(0,0) = (intensity_ ((*input_) (1,0)) - intensity_ ((*input_) (0,0))) * 0.5;

// #ifdef _OPENMP
// //#pragma omp parallel for shared (derivatives_cols_, input_) num_threads (threads_)
// #pragma omp parallel for num_threads (threads_)
// #endif
  for(int i = 1; i < w; ++i)
	{
		derivatives_cols_(i,0) = (intensity_ ((*input_) (i,1)) - intensity_ ((*input_) (i,0))) * 0.5;
	}

  derivatives_rows_(w,0) = (intensity_ ((*input_) (w,0)) - intensity_ ((*input_) (w-1,0))) * 0.5;
  derivatives_cols_(w,0) = (intensity_ ((*input_) (w,1)) - intensity_ ((*input_) (w,0))) * 0.5;

// #ifdef _OPENMP
// //#pragma omp parallel for shared (derivatives_cols_, derivatives_rows_, input_) num_threads (threads_)
// #pragma omp parallel for num_threads (threads_)
// #endif
  for(int j = 1; j < h; ++j)
  {
    // i = 0 --> i-1 out of range ; use 0
		derivatives_rows_(0,j) = (intensity_ ((*input_) (1,j)) - intensity_ ((*input_) (0,j))) * 0.5;
    for(int i = 1; i < w; ++i)
    {
      // derivative with respect to rows
      derivatives_rows_(i,j) = (intensity_ ((*input_) (i+1,j)) - intensity_ ((*input_) (i-1,j))) * 0.5;

      // derivative with respect to cols
      derivatives_cols_(i,j) = (intensity_ ((*input_) (i,j+1)) - intensity_ ((*input_) (i,j-1))) * 0.5;
    }
    // i = w --> w+1 out of range ; use w
    derivatives_rows_(w,j) = (intensity_ ((*input_) (w,j)) - intensity_ ((*input_) (w-1,j))) * 0.5;
  }

  // j = h --> j+1 out of range use h
  derivatives_cols_(0,h) = (intensity_ ((*input_) (0,h)) - intensity_ ((*input_) (0,h-1))) * 0.5;
  derivatives_rows_(0,h) = (intensity_ ((*input_) (1,h)) - intensity_ ((*input_) (0,h))) * 0.5;

// #ifdef _OPENMP
// //#pragma omp parallel for shared (derivatives_cols_, input_) num_threads (threads_)
// #pragma omp parallel for num_threads (threads_)
// #endif
  for(int i = 1; i < w; ++i)
	{
    derivatives_cols_(i,h) = (intensity_ ((*input_) (i,h)) - intensity_ ((*input_) (i,h-1))) * 0.5;
	}
  derivatives_rows_(w,h) = (intensity_ ((*input_) (w,h)) - intensity_ ((*input_) (w-1,h))) * 0.5;
  derivatives_cols_(w,h) = (intensity_ ((*input_) (w,h)) - intensity_ ((*input_) (w,h-1))) * 0.5;

  float highest_response_;
  
  switch (method_)
  {
    case HARRIS:
      responseHarris(*response_, highest_response_);
      break;
    case NOBLE:
      responseNoble(*response_, highest_response_);
      break;
    case LOWE:
      responseLowe(*response_, highest_response_);
      break;
    case TOMASI:
      responseTomasi(*response_, highest_response_);
      break;
  }
  
  if (!nonmax_)
    output = *response_;
  else
  {    
    threshold_*= highest_response_;

    std::sort (indices_->begin (), indices_->end (), 
               boost::bind (&HarrisKeypoint2D::greaterIntensityAtIndices, this, _1, _2));
    
    output.clear ();
    output.reserve (response_->size());
    std::vector<bool> occupency_map (response_->size (), false);    
    int width (response_->width);
    int height (response_->height);
    const int occupency_map_size (occupency_map.size ());

#ifdef _OPENMP
#pragma omp parallel for shared (output, occupency_map) private (width, height) num_threads(threads_)   
#endif
    for (int idx = 0; idx < occupency_map_size; ++idx)
    {
//.........这里部分代码省略.........