iss_3d.hpp

/*
 * Software License Agreement (BSD License)
 *
 *  Point Cloud Library (PCL) - www.pointclouds.org
 *  Copyright (c) 2010-2011, Willow Garage, Inc.
 *
 *  All rights reserved.
 *
 *  Redistribution and use in source and binary forms, with or without
 *  modification, are permitted provided that the following conditions
 *  are met:
 *
 *   * Redistributions of source code must retain the above copyright
 *     notice, this list of conditions and the following disclaimer.
 *   * Redistributions in binary form must reproduce the above
 *     copyright notice, this list of conditions and the following
 *     disclaimer in the documentation and/or other materials provided
 *     with the distribution.
 *   * Neither the name of Willow Garage, Inc. nor the names of its
 *     contributors may be used to endorse or promote products derived
 *     from this software without specific prior written permission.
 *
 *  THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
 *  "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
 *  LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
 *  FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
 *  COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
 *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
 *  BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
 *  LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
 *  CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 *  LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
 *  ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
 *  POSSIBILITY OF SUCH DAMAGE.
 *
 */
 
#ifndef PCL_ISS_KEYPOINT3D_IMPL_H_
#define PCL_ISS_KEYPOINT3D_IMPL_H_
 
#include <Eigen/Eigenvalues> // for SelfAdjointEigenSolver
#include <pcl/features/boundary.h>
#include <pcl/features/normal_3d.h>
#include <pcl/features/integral_image_normal.h>
 
#include <pcl/keypoints/iss_3d.h>
 
//////////////////////////////////////////////////////////////////////////////////////////////
template<typename PointInT, typename PointOutT, typename NormalT> void
pcl::ISSKeypoint3D<PointInT, PointOutT, NormalT>::setSalientRadius (double salient_radius)
{
  salient_radius_ = salient_radius;
}
 
//////////////////////////////////////////////////////////////////////////////////////////////
template<typename PointInT, typename PointOutT, typename NormalT> void
pcl::ISSKeypoint3D<PointInT, PointOutT, NormalT>::setNonMaxRadius (double non_max_radius)
{
  non_max_radius_ = non_max_radius;
}
 
//////////////////////////////////////////////////////////////////////////////////////////////
template<typename PointInT, typename PointOutT, typename NormalT> void
pcl::ISSKeypoint3D<PointInT, PointOutT, NormalT>::setNormalRadius (double normal_radius)
{
  normal_radius_ = normal_radius;
}
 
//////////////////////////////////////////////////////////////////////////////////////////////
template<typename PointInT, typename PointOutT, typename NormalT> void
pcl::ISSKeypoint3D<PointInT, PointOutT, NormalT>::setBorderRadius (double border_radius)
{
  border_radius_ = border_radius;
}
 
//////////////////////////////////////////////////////////////////////////////////////////////
template<typename PointInT, typename PointOutT, typename NormalT> void
pcl::ISSKeypoint3D<PointInT, PointOutT, NormalT>::setThreshold21 (double gamma_21)
{
  gamma_21_ = gamma_21;
}
 
//////////////////////////////////////////////////////////////////////////////////////////////
template<typename PointInT, typename PointOutT, typename NormalT> void
pcl::ISSKeypoint3D<PointInT, PointOutT, NormalT>::setThreshold32 (double gamma_32)
{
  gamma_32_ = gamma_32;
}
 
//////////////////////////////////////////////////////////////////////////////////////////////
template<typename PointInT, typename PointOutT, typename NormalT> void
pcl::ISSKeypoint3D<PointInT, PointOutT, NormalT>::setMinNeighbors (int min_neighbors)
{
  min_neighbors_ = min_neighbors;
}
 
//////////////////////////////////////////////////////////////////////////////////////////////
template<typename PointInT, typename PointOutT, typename NormalT> void
pcl::ISSKeypoint3D<PointInT, PointOutT, NormalT>::setNormals (const PointCloudNConstPtr &normals)
{
  normals_ = normals;
}
 
//////////////////////////////////////////////////////////////////////////////////////////////
template<typename PointInT, typename PointOutT, typename NormalT> void
pcl::ISSKeypoint3D<PointInT, PointOutT, NormalT>::setNumberOfThreads (unsigned int nr_threads)
{
  if (nr_threads == 0)
#ifdef _OPENMP
    threads_ = omp_get_num_procs();
#else
    threads_ = 1;
#endif
  else
    threads_ = nr_threads;
}
 
//////////////////////////////////////////////////////////////////////////////////////////////
template<typename PointInT, typename PointOutT, typename NormalT> bool*
pcl::ISSKeypoint3D<PointInT, PointOutT, NormalT>::getBoundaryPoints (PointCloudIn &input, double border_radius, float angle_threshold)
{
  bool* edge_points = new bool [input.size ()];
 
  Eigen::Vector4f u = Eigen::Vector4f::Zero ();
  Eigen::Vector4f v = Eigen::Vector4f::Zero ();
 
  pcl::BoundaryEstimation<PointInT, NormalT, pcl::Boundary> boundary_estimator;
  boundary_estimator.setInputCloud (input_);
 
#pragma omp parallel for \
  default(none) \
  shared(angle_threshold, boundary_estimator, border_radius, edge_points, input) \
  firstprivate(u, v) \
  num_threads(threads_)
  for (int index = 0; index < int (input.size ()); index++)
  {
    edge_points[index] = false;
    PointInT current_point = input[index];
 
    if (pcl::isFinite(current_point))
    {
      pcl::Indices nn_indices;
      std::vector<float> nn_distances;
      int n_neighbors;
 
      this->searchForNeighbors (index, border_radius, nn_indices, nn_distances);
 
      n_neighbors = static_cast<int> (nn_indices.size ());
 
      if (n_neighbors >= min_neighbors_)
      {
  boundary_estimator.getCoordinateSystemOnPlane ((*normals_)[index], u, v);
 
  if (boundary_estimator.isBoundaryPoint (input, static_cast<int> (index), nn_indices, u, v, angle_threshold))
    edge_points[index] = true;
      }
    }
  }
 
  return (edge_points);
}
 
//////////////////////////////////////////////////////////////////////////////////////////////
template<typename PointInT, typename PointOutT, typename NormalT> void
pcl::ISSKeypoint3D<PointInT, PointOutT, NormalT>::getScatterMatrix (const int& current_index, Eigen::Matrix3d &cov_m)
{
  const PointInT& current_point = (*input_)[current_index];
 
  double central_point[3];
  memset(central_point, 0, sizeof(double) * 3);
 
  central_point[0] = current_point.x;
  central_point[1] = current_point.y;
  central_point[2] = current_point.z;
 
  cov_m = Eigen::Matrix3d::Zero ();
 
  pcl::Indices nn_indices;
  std::vector<float> nn_distances;
  int n_neighbors;
 
  this->searchForNeighbors (current_index, salient_radius_, nn_indices, nn_distances);
 
  n_neighbors = static_cast<int> (nn_indices.size ());
 
  if (n_neighbors < min_neighbors_)
    return;
 
  double cov[9];
  memset(cov, 0, sizeof(double) * 9);
 
  for (const auto& n_idx : nn_indices)
  {
    const PointInT& n_point = (*input_)[n_idx];
 
    double neigh_point[3];
    memset(neigh_point, 0, sizeof(double) * 3);
 
    neigh_point[0] = n_point.x;
    neigh_point[1] = n_point.y;
    neigh_point[2] = n_point.z;
 
    for (int i = 0; i < 3; i++)
      for (int j = 0; j < 3; j++)
        cov[i * 3 + j] += (neigh_point[i] - central_point[i]) * (neigh_point[j] - central_point[j]);
  }
 
  cov_m << cov[0], cov[1], cov[2],
     cov[3], cov[4], cov[5],
     cov[6], cov[7], cov[8];
}
 
//////////////////////////////////////////////////////////////////////////////////////////////
template<typename PointInT, typename PointOutT, typename NormalT> bool
pcl::ISSKeypoint3D<PointInT, PointOutT, NormalT>::initCompute ()
{
  if (!Keypoint<PointInT, PointOutT>::initCompute ())
  {
    PCL_ERROR ("[pcl::%s::initCompute] init failed!\n", name_.c_str ());
    return (false);
  }
  if (salient_radius_ <= 0)
  {
    PCL_ERROR ("[pcl::%s::initCompute] : the salient radius (%f) must be strict positive!\n",
    name_.c_str (), salient_radius_);
    return (false);
  }
  if (non_max_radius_ <= 0)
  {
    PCL_ERROR ("[pcl::%s::initCompute] : the non maxima radius (%f) must be strict positive!\n",
    name_.c_str (), non_max_radius_);
    return (false);
  }
  if (gamma_21_ <= 0)
  {
    PCL_ERROR ("[pcl::%s::initCompute] : the threshold on the ratio between the 2nd and the 1rst eigenvalue (%f) must be strict positive!\n",
    name_.c_str (), gamma_21_);
    return (false);
  }
  if (gamma_32_ <= 0)
  {
    PCL_ERROR ("[pcl::%s::initCompute] : the threshold on the ratio between the 3rd and the 2nd eigenvalue (%f) must be strict positive!\n",
    name_.c_str (), gamma_32_);
    return (false);
  }
  if (min_neighbors_ <= 0)
  {
    PCL_ERROR ("[pcl::%s::initCompute] : the minimum number of neighbors (%f) must be strict positive!\n",
    name_.c_str (), min_neighbors_);
    return (false);
  }
 
    delete[] third_eigen_value_;
 
  third_eigen_value_ = new double[input_->size ()];
  memset(third_eigen_value_, 0, sizeof(double) * input_->size ());
 
    delete[] edge_points_;
 
  if (border_radius_ > 0.0)
  {
    if (normals_->empty ())
    {
      if (normal_radius_ <= 0.)
      {
        PCL_ERROR ("[pcl::%s::initCompute] : the radius used to estimate surface normals (%f) must be positive!\n",
        name_.c_str (), normal_radius_);
        return (false);
      }
 
      PointCloudNPtr normal_ptr (new PointCloudN ());
      if (input_->height == 1 )
      {
        pcl::NormalEstimation<PointInT, NormalT> normal_estimation;
        normal_estimation.setInputCloud (surface_);
        normal_estimation.setRadiusSearch (normal_radius_);
        normal_estimation.compute (*normal_ptr);
      }
      else
      {
        pcl::IntegralImageNormalEstimation<PointInT, NormalT> normal_estimation;
        normal_estimation.setNormalEstimationMethod (pcl::IntegralImageNormalEstimation<PointInT, NormalT>::SIMPLE_3D_GRADIENT);
        normal_estimation.setInputCloud (surface_);
        normal_estimation.setNormalSmoothingSize (5.0);
        normal_estimation.compute (*normal_ptr);
      }
      normals_ = normal_ptr;
    }
    if (normals_->size () != surface_->size ())
    {
      PCL_ERROR ("[pcl::%s::initCompute] normals given, but the number of normals does not match the number of input points!\n", name_.c_str ());
      return (false);
    }
  }
  else if (border_radius_ < 0.0)
  {
    PCL_ERROR ("[pcl::%s::initCompute] : the border radius used to estimate boundary points (%f) must be positive!\n",
    name_.c_str (), border_radius_);
    return (false);
  }
 
  return (true);
}
 
//////////////////////////////////////////////////////////////////////////////////////////////
template<typename PointInT, typename PointOutT, typename NormalT> void
pcl::ISSKeypoint3D<PointInT, PointOutT, NormalT>::detectKeypoints (PointCloudOut &output)
{
  // Make sure the output cloud is empty
  output.clear ();
 
  if (border_radius_ > 0.0)
    edge_points_ = getBoundaryPoints (*(input_->makeShared ()), border_radius_, angle_threshold_);
 
  bool* borders = new bool [input_->size()];
 
#pragma omp parallel for \
  default(none) \
  shared(borders) \
  num_threads(threads_)
  for (int index = 0; index < int (input_->size ()); index++)
  {
    borders[index] = false;
    PointInT current_point = (*input_)[index];
 
    if ((border_radius_ > 0.0) && (pcl::isFinite(current_point)))
    {
      pcl::Indices nn_indices;
      std::vector<float> nn_distances;
 
      this->searchForNeighbors (index, border_radius_, nn_indices, nn_distances);
 
      for (const auto &nn_index : nn_indices)
      {
        if (edge_points_[nn_index])
        {
          borders[index] = true;
          break;
        }
      }
    }
  }
 
#ifdef _OPENMP
  Eigen::Vector3d *omp_mem = new Eigen::Vector3d[threads_];
 
  for (std::size_t i = 0; i < threads_; i++)
    omp_mem[i].setZero (3);
#else
  Eigen::Vector3d *omp_mem = new Eigen::Vector3d[1];
 
  omp_mem[0].setZero (3);
#endif
 
  double *prg_local_mem = new double[input_->size () * 3];
  double **prg_mem = new double * [input_->size ()];
 
  for (std::size_t i = 0; i < input_->size (); i++)
    prg_mem[i] = prg_local_mem + 3 * i;
 
#pragma omp parallel for \
  default(none) \
  shared(borders, omp_mem, prg_mem) \
  num_threads(threads_)
  for (int index = 0; index < static_cast<int> (input_->size ()); index++)
  {
#ifdef _OPENMP
    int tid = omp_get_thread_num ();
#else
    int tid = 0;
#endif
    PointInT current_point = (*input_)[index];
 
    if ((!borders[index]) && pcl::isFinite(current_point))
    {
      //if the considered point is not a border point and the point is "finite", then compute the scatter matrix
      Eigen::Matrix3d cov_m = Eigen::Matrix3d::Zero ();
      getScatterMatrix (static_cast<int> (index), cov_m);
 
      Eigen::SelfAdjointEigenSolver<Eigen::Matrix3d> solver (cov_m);
 
      const double& e1c = solver.eigenvalues ()[2];
      const double& e2c = solver.eigenvalues ()[1];
      const double& e3c = solver.eigenvalues ()[0];
 
      if (!std::isfinite (e1c) || !std::isfinite (e2c) || !std::isfinite (e3c))
  continue;
 
      if (e3c < 0)
      {
  PCL_WARN ("[pcl::%s::detectKeypoints] : The third eigenvalue is negative! Skipping the point with index %i.\n",
            name_.c_str (), index);
  continue;
      }
 
      omp_mem[tid][0] = e2c / e1c;
      omp_mem[tid][1] = e3c / e2c;;
      omp_mem[tid][2] = e3c;
    }
 
    for (Eigen::Index d = 0; d < omp_mem[tid].size (); d++)
        prg_mem[index][d] = omp_mem[tid][d];
  }
 
  for (int index = 0; index < int (input_->size ()); index++)
  {
   if (!borders[index])
    {
      if ((prg_mem[index][0] < gamma_21_) && (prg_mem[index][1] < gamma_32_))
        third_eigen_value_[index] = prg_mem[index][2];
    }
  }
 
  bool* feat_max = new bool [input_->size()];
 
#pragma omp parallel for \
  default(none) \
  shared(feat_max) \
  num_threads(threads_)
  for (int index = 0; index < int (input_->size ()); index++)
  {
    feat_max [index] = false;
    PointInT current_point = (*input_)[index];
 
    if ((third_eigen_value_[index] > 0.0) && (pcl::isFinite(current_point)))
    {
      pcl::Indices nn_indices;
      std::vector<float> nn_distances;
      int n_neighbors;
 
      this->searchForNeighbors (index, non_max_radius_, nn_indices, nn_distances);
 
      n_neighbors = static_cast<int> (nn_indices.size ());
 
      if (n_neighbors >= min_neighbors_)
      {
        bool is_max = true;
 
        for (const auto& j : nn_indices)
          if (third_eigen_value_[index] < third_eigen_value_[j])
            is_max = false;
        if (is_max)
          feat_max[index] = true;
      }
    }
  }
 
#pragma omp parallel for \
  default(none) \
  shared(feat_max, output) \
  num_threads(threads_)
  for (int index = 0; index < int (input_->size ()); index++)
  {
    if (feat_max[index])
#pragma omp critical
    {
      PointOutT p;
      p.getVector3fMap () = (*input_)[index].getVector3fMap ();
      output.push_back(p);
      keypoints_indices_->indices.push_back (index);
    }
  }
 
  output.header = input_->header;
  output.width = output.size ();
  output.height = 1;
 
  // Clear the contents of variables and arrays before the beginning of the next computation.
  if (border_radius_ > 0.0)
    normals_.reset (new pcl::PointCloud<NormalT>);
 
  delete[] borders;
  delete[] prg_mem;
  delete[] prg_local_mem;
  delete[] feat_max;
  delete[] omp_mem;
}
 
#define PCL_INSTANTIATE_ISSKeypoint3D(T,U,N) template class PCL_EXPORTS pcl::ISSKeypoint3D<T,U,N>;
 
#endif /* PCL_ISS_3D_IMPL_H_ */