search.h

/*
 * Software License Agreement (BSD License)
 *
 *  Point Cloud Library (PCL) - www.pointclouds.org
 *  Copyright (c) 2010-2011, Willow Garage, Inc.
 *  Copyright (c) 2012-, Open Perception, 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 the copyright holder(s) 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.
 *
 */
 
#pragma once
 
#include <pcl/pcl_base.h> // for IndicesConstPtr
#include <pcl/point_cloud.h>
#include <pcl/for_each_type.h>
#include <pcl/common/concatenate.h>
#include <pcl/common/copy_point.h>
 
namespace pcl
{
  // Forward declarations
  template <typename T> class PointRepresentation;
 
  namespace search
  {
    /** \brief Generic search class. All search wrappers must inherit from this.
      *
      * Each search method must implement 2 different types of search:
      *   - \b nearestKSearch - search for K-nearest neighbors.
      *   - \b radiusSearch - search for all nearest neighbors in a sphere of a given radius
      *
      * The input to each search method can be given in 3 different ways:
      *   - as a query point
      *   - as a (cloud, index) pair
      *   - as an index
      *
      * For the latter option, it is assumed that the user specified the input
      * via a \ref setInputCloud () method first.
      *
      * \note In case of an error, all methods are supposed to return 0 as the number of neighbors found.
      *
      * \note libpcl_search deals with three-dimensional search problems. For higher
      * level dimensional search, please refer to the libpcl_kdtree module.
      *
      * \author Radu B. Rusu
      * \ingroup search
      */
    template<typename PointT>
    class Search
    {
      public:
        using PointCloud = pcl::PointCloud<PointT>;
        using PointCloudPtr = typename PointCloud::Ptr;
        using PointCloudConstPtr = typename PointCloud::ConstPtr;
 
        using Ptr = shared_ptr<pcl::search::Search<PointT> >;
        using ConstPtr = shared_ptr<const pcl::search::Search<PointT> >;
 
        using IndicesPtr = pcl::IndicesPtr;
        using IndicesConstPtr = pcl::IndicesConstPtr;
 
        using PointRepresentationConstPtr = shared_ptr<const PointRepresentation<PointT> >;
 
        /** Constructor. */
        Search (const std::string& name = "", bool sorted = false);
 
        /** Destructor. */
        virtual
        ~Search () = default;
 
        /** \brief Returns the search method name
          */
        virtual const std::string& 
        getName () const;
 
        /** \brief sets whether the results should be sorted (ascending in the distance) or not
          * \param[in] sorted should be true if the results should be sorted by the distance in ascending order.
          * Otherwise the results may be returned in any order.
          */
        virtual void 
        setSortedResults (bool sorted);
 
        /** \brief Gets whether the results should be sorted (ascending in the distance) or not
          * Otherwise the results may be returned in any order.
          */
        virtual bool 
        getSortedResults ();
 
        /** \brief Provide a pointer to the point representation to use to convert points into
         * k-D vectors. If you want to use this function, it is recommended to do so _before_
         * calling setInputCloud. Some search methods will ignore this (e.g. octree or organized search).
         * \param[in] point_representation the const shared pointer to a PointRepresentation
         * \return True if success, false if a problem occurred or the search method will not use the point representation
         */
        virtual bool
        setPointRepresentation (const PointRepresentationConstPtr&)
        {
          return (false);
        }
        
        /** \brief Pass the input dataset that the search will be performed on.
          * \param[in] cloud a const pointer to the PointCloud data
          * \param[in] indices the point indices subset that is to be used from the cloud
          * \return True if successful, false if an error occurred, for example because the point cloud is unsuited for the search method.
          */
        virtual bool
        setInputCloud (const PointCloudConstPtr& cloud, 
                       const IndicesConstPtr &indices = IndicesConstPtr ());
 
        /** \brief Get a pointer to the input point cloud dataset. */
        virtual PointCloudConstPtr
        getInputCloud () const
        {
          return (input_);
        }
 
        /** \brief Get a pointer to the vector of indices used. */
        virtual IndicesConstPtr
        getIndices () const
        {
          return (indices_);
        }
 
        /** \brief Search for the k-nearest neighbors for the given query point.
          * \param[in] point the given query point
          * \param[in] k the number of neighbors to search for
          * \param[out] k_indices the resultant indices of the neighboring points (must be resized to \a k a priori!)
          * \param[out] k_sqr_distances the resultant squared distances to the neighboring points (must be resized to \a k
          * a priori!)
          * \return number of neighbors found
          */
        virtual int
        nearestKSearch (const PointT &point, int k, Indices &k_indices,
                        std::vector<float> &k_sqr_distances) const = 0;
 
        /** \brief Search for k-nearest neighbors for the given query point.
          * This method accepts a different template parameter for the point type.
          * \param[in] point the given query point
          * \param[in] k the number of neighbors to search for
          * \param[out] k_indices the resultant indices of the neighboring points (must be resized to \a k a priori!)
          * \param[out] k_sqr_distances the resultant squared distances to the neighboring points (must be resized to \a k
          * a priori!)
          * \return number of neighbors found
          */
        template <typename PointTDiff> inline int
        nearestKSearchT (const PointTDiff &point, int k,
                         Indices &k_indices, std::vector<float> &k_sqr_distances) const
        {
          PointT p;
          copyPoint (point, p);
          return (nearestKSearch (p, k, k_indices, k_sqr_distances));
        }
 
        /** \brief Search for k-nearest neighbors for the given query point.
          *
          * \attention This method does not do any bounds checking for the input index
          * (i.e., index >= cloud.size () || index < 0), and assumes valid (i.e., finite) data.
          *
          * \param[in] cloud the point cloud data
          * \param[in] index a \a valid index in \a cloud representing a \a valid (i.e., finite) query point
          * \param[in] k the number of neighbors to search for
          * \param[out] k_indices the resultant indices of the neighboring points (must be resized to \a k a priori!)
          * \param[out] k_sqr_distances the resultant squared distances to the neighboring points (must be resized to \a k
          * a priori!)
          *
          * \return number of neighbors found
          *
          * \exception asserts in debug mode if the index is not between 0 and the maximum number of points
          */
        virtual int
        nearestKSearch (const PointCloud &cloud, index_t index, int k,
                        Indices &k_indices,
                        std::vector<float> &k_sqr_distances) const;
 
        /** \brief Search for k-nearest neighbors for the given query point (zero-copy).
          *
          * \attention This method does not do any bounds checking for the input index
          * (i.e., index >= cloud.size () || index < 0), and assumes valid (i.e., finite) data.
          *
          * \param[in] index a \a valid index representing a \a valid query point in the dataset given
          * by \a setInputCloud. If indices were given in setInputCloud, index will be the position in
          * the indices vector.
          *
          * \param[in] k the number of neighbors to search for
          * \param[out] k_indices the resultant indices of the neighboring points (must be resized to \a k a priori!)
          * \param[out] k_sqr_distances the resultant squared distances to the neighboring points (must be resized to \a k
          * a priori!)
          * \return number of neighbors found
          *
          * \exception asserts in debug mode if the index is not between 0 and the maximum number of points
          */
        virtual int
        nearestKSearch (index_t index, int k,
                        Indices &k_indices,
                        std::vector<float> &k_sqr_distances) const;
 
        /** \brief Search for the k-nearest neighbors for the given query point.
          * \param[in] cloud the point cloud data
          * \param[in] indices a vector of point cloud indices to query for nearest neighbors
          * \param[in] k the number of neighbors to search for
          * \param[out] k_indices the resultant indices of the neighboring points, k_indices[i] corresponds to the neighbors of the query point i
          * \param[out] k_sqr_distances the resultant squared distances to the neighboring points, k_sqr_distances[i] corresponds to the neighbors of the query point i
          */
        virtual void
        nearestKSearch (const PointCloud& cloud, const Indices& indices,
                        int k, std::vector<Indices>& k_indices,
                        std::vector< std::vector<float> >& k_sqr_distances) const;
 
        /** \brief Search for the k-nearest neighbors for the given query point. Use this method if the query points are of a different type than the points in the data set (e.g. PointXYZRGBA instead of PointXYZ).
          * \param[in] cloud the point cloud data
          * \param[in] indices a vector of point cloud indices to query for nearest neighbors
          * \param[in] k the number of neighbors to search for
          * \param[out] k_indices the resultant indices of the neighboring points, k_indices[i] corresponds to the neighbors of the query point i
          * \param[out] k_sqr_distances the resultant squared distances to the neighboring points, k_sqr_distances[i] corresponds to the neighbors of the query point i
          * \note This method copies the input point cloud of type PointTDiff to a temporary cloud of type PointT and performs the batch search on the new cloud. You should prefer the single-point search if you don't use a search algorithm that accelerates batch NN search.
          */
        template <typename PointTDiff> void
        nearestKSearchT (const pcl::PointCloud<PointTDiff> &cloud, const Indices& indices, int k, std::vector<Indices> &k_indices,
                         std::vector< std::vector<float> > &k_sqr_distances) const
        {
          // Copy all the data fields from the input cloud to the output one
          using FieldListInT = typename pcl::traits::fieldList<PointT>::type;
          using FieldListOutT = typename pcl::traits::fieldList<PointTDiff>::type;
          using FieldList = typename pcl::intersect<FieldListInT, FieldListOutT>::type;
 
          pcl::PointCloud<PointT> pc;
          if (indices.empty ())
          {
            pc.resize (cloud.size());
            for (std::size_t i = 0; i < cloud.size(); i++)
            {
              pcl::for_each_type <FieldList> (pcl::NdConcatenateFunctor <PointTDiff, PointT> (
                                              cloud[i], pc[i]));
            }
            nearestKSearch (pc,Indices(),k,k_indices,k_sqr_distances);
          }
          else
          {
            pc.resize (indices.size());
            for (std::size_t i = 0; i < indices.size(); i++)
            {
              pcl::for_each_type <FieldList> (pcl::NdConcatenateFunctor <PointTDiff, PointT> (
                                              cloud[indices[i]], pc[i]));
            }
            nearestKSearch (pc,Indices(),k,k_indices,k_sqr_distances);
          }
        }
 
        /** \brief Search for all the nearest neighbors of the query point in a given radius.
          * \param[in] point the given query point
          * \param[in] radius the radius of the sphere bounding all of p_q's neighbors
          * \param[out] k_indices the resultant indices of the neighboring points
          * \param[out] k_sqr_distances the resultant squared distances to the neighboring points
          * \param[in] max_nn if given, bounds the maximum returned neighbors to this value. If \a max_nn is set to
          * 0 or to a number higher than the number of points in the input cloud, all neighbors in \a radius will be
          * returned.
          * \return number of neighbors found in radius
          */
        virtual int
        radiusSearch (const PointT& point, double radius, Indices& k_indices,
                      std::vector<float>& k_sqr_distances, unsigned int max_nn = 0) const = 0;
 
        /** \brief Search for all the nearest neighbors of the query point in a given radius.
          * \param[in] point the given query point
          * \param[in] radius the radius of the sphere bounding all of p_q's neighbors
          * \param[out] k_indices the resultant indices of the neighboring points
          * \param[out] k_sqr_distances the resultant squared distances to the neighboring points
          * \param[in] max_nn if given, bounds the maximum returned neighbors to this value. If \a max_nn is set to
          * 0 or to a number higher than the number of points in the input cloud, all neighbors in \a radius will be
          * returned.
          * \return number of neighbors found in radius
          */
        template <typename PointTDiff> inline int
        radiusSearchT (const PointTDiff &point, double radius, Indices &k_indices,
                       std::vector<float> &k_sqr_distances, unsigned int max_nn = 0) const
        {
          PointT p;
          copyPoint (point, p);
          return (radiusSearch (p, radius, k_indices, k_sqr_distances, max_nn));
        }
 
        /** \brief Search for all the nearest neighbors of the query point in a given radius.
          *
          * \attention This method does not do any bounds checking for the input index
          * (i.e., index >= cloud.size () || index < 0), and assumes valid (i.e., finite) data.
          *
          * \param[in] cloud the point cloud data
          * \param[in] index a \a valid index in \a cloud representing a \a valid (i.e., finite) query point
          * \param[in] radius the radius of the sphere bounding all of p_q's neighbors
          * \param[out] k_indices the resultant indices of the neighboring points
          * \param[out] k_sqr_distances the resultant squared distances to the neighboring points
          * \param[in] max_nn if given, bounds the maximum returned neighbors to this value. If \a max_nn is set to
          * 0 or to a number higher than the number of points in the input cloud, all neighbors in \a radius will be
          * returned.
          * \return number of neighbors found in radius
          *
          * \exception asserts in debug mode if the index is not between 0 and the maximum number of points
          */
        virtual int
        radiusSearch (const PointCloud &cloud, index_t index, double radius,
                      Indices &k_indices, std::vector<float> &k_sqr_distances,
                      unsigned int max_nn = 0) const;
 
        /** \brief Search for all the nearest neighbors of the query point in a given radius (zero-copy).
          *
          * \attention This method does not do any bounds checking for the input index
          * (i.e., index >= cloud.size () || index < 0), and assumes valid (i.e., finite) data.
          *
          * \param[in] index a \a valid index representing a \a valid query point in the dataset given
          * by \a setInputCloud. If indices were given in setInputCloud, index will be the position in
          * the indices vector.
          *
          * \param[in] radius the radius of the sphere bounding all of p_q's neighbors
          * \param[out] k_indices the resultant indices of the neighboring points
          * \param[out] k_sqr_distances the resultant squared distances to the neighboring points
          * \param[in] max_nn if given, bounds the maximum returned neighbors to this value. If \a max_nn is set to
          * 0 or to a number higher than the number of points in the input cloud, all neighbors in \a radius will be
          * returned.
          * \return number of neighbors found in radius
          *
          * \exception asserts in debug mode if the index is not between 0 and the maximum number of points
          */
        virtual int
        radiusSearch (index_t index, double radius, Indices &k_indices,
                      std::vector<float> &k_sqr_distances, unsigned int max_nn = 0) const;
 
        /** \brief Search for all the nearest neighbors of the query point in a given radius.
          * \param[in] cloud the point cloud data
          * \param[in] indices the indices in \a cloud. If indices is empty, neighbors will be searched for all points.
          * \param[in] radius the radius of the sphere bounding all of p_q's neighbors
          * \param[out] k_indices the resultant indices of the neighboring points, k_indices[i] corresponds to the neighbors of the query point i
          * \param[out] k_sqr_distances the resultant squared distances to the neighboring points, k_sqr_distances[i] corresponds to the neighbors of the query point i
          * \param[in] max_nn if given, bounds the maximum returned neighbors to this value. If \a max_nn is set to
          * 0 or to a number higher than the number of points in the input cloud, all neighbors in \a radius will be
          * returned.
          */
        virtual void
        radiusSearch (const PointCloud& cloud,
                      const Indices& indices,
                      double radius,
                      std::vector<Indices>& k_indices,
                      std::vector< std::vector<float> > &k_sqr_distances,
                      unsigned int max_nn = 0) const;
 
        /** \brief Search for all the nearest neighbors of the query points in a given radius.
          * \param[in] cloud the point cloud data
          * \param[in] indices a vector of point cloud indices to query for nearest neighbors
          * \param[in] radius the radius of the sphere bounding all of p_q's neighbors
          * \param[out] k_indices the resultant indices of the neighboring points, k_indices[i] corresponds to the neighbors of the query point i
          * \param[out] k_sqr_distances the resultant squared distances to the neighboring points, k_sqr_distances[i] corresponds to the neighbors of the query point i
          * \param[in] max_nn if given, bounds the maximum returned neighbors to this value. If \a max_nn is set to
          * 0 or to a number higher than the number of points in the input cloud, all neighbors in \a radius will be
          * returned.
          * \note This method copies the input point cloud of type PointTDiff to a temporary cloud of type PointT and performs the batch search on the new cloud. You should prefer the single-point search if you don't use a search algorithm that accelerates batch NN search.
          */
        template <typename PointTDiff> void
        radiusSearchT (const pcl::PointCloud<PointTDiff> &cloud,
                       const Indices& indices,
                       double radius,
                       std::vector<Indices> &k_indices,
                       std::vector< std::vector<float> > &k_sqr_distances,
                       unsigned int max_nn = 0) const
        {
          // Copy all the data fields from the input cloud to the output one
          using FieldListInT = typename pcl::traits::fieldList<PointT>::type;
          using FieldListOutT = typename pcl::traits::fieldList<PointTDiff>::type;
          using FieldList = typename pcl::intersect<FieldListInT, FieldListOutT>::type;
 
          pcl::PointCloud<PointT> pc;
          if (indices.empty ())
          {
            pc.resize (cloud.size ());
            for (std::size_t i = 0; i < cloud.size (); ++i)
              pcl::for_each_type <FieldList> (pcl::NdConcatenateFunctor <PointTDiff, PointT> (cloud[i], pc[i]));
            radiusSearch (pc, Indices(), radius, k_indices, k_sqr_distances, max_nn);
          }
          else
          {
            pc.resize (indices.size ());
            for (std::size_t i = 0; i < indices.size (); ++i)
              pcl::for_each_type <FieldList> (pcl::NdConcatenateFunctor <PointTDiff, PointT> (cloud[indices[i]], pc[i]));
            radiusSearch (pc, Indices(), radius, k_indices, k_sqr_distances, max_nn);
          }
        }
 
        /** \brief Set the number of threads to use for searching over multiple points or indices
         * \param[in] nr_threads the number of threads to use (0 automatically sets the threads based on the hardware)
         */
        void setNumberOfThreads(unsigned int nr_threads) {
          #ifdef _OPENMP
          num_threads_ = nr_threads != 0 ? nr_threads : omp_get_num_procs();
          #else
          if (nr_threads != 1) {
            PCL_WARN("OpenMP is not available. Keeping number of threads unchanged at 1\n");
          }
          #endif
        }
 
      protected:
        void 
        sortResults (Indices& indices, std::vector<float>& distances) const;
 
        PointCloudConstPtr input_;
        IndicesConstPtr indices_;
        bool sorted_results_;
        std::string name_;
 
        /** \brief The number of threads to use when searching over multiple points or indices */
        unsigned int num_threads_{1};
        
      private:
        struct Compare
        {
          Compare (const std::vector<float>& distances)
          : distances_ (distances)
          {
          }
          
          bool 
          operator () (index_t first, index_t second) const
          {
            return (distances_ [first] < distances_[second]);
          }
 
          const std::vector<float>& distances_;
        };
    }; // class Search    
  } // namespace search
} // namespace pcl
 
#ifdef PCL_NO_PRECOMPILE
#include <pcl/search/impl/search.hpp>
#endif