41 #ifndef PCL_SAMPLE_CONSENSUS_IMPL_SAC_MODEL_SPHERE_H_
42 #define PCL_SAMPLE_CONSENSUS_IMPL_SAC_MODEL_SPHERE_H_
44 #include <pcl/sample_consensus/eigen.h>
45 #include <pcl/sample_consensus/sac_model_sphere.h>
48 template <
typename Po
intT>
bool
51 if (samples.size () != sample_size_)
53 PCL_ERROR (
"[pcl::SampleConsensusModelSphere::isSampleGood] Wrong number of samples (is %lu, should be %lu)!\n", samples.size (), sample_size_);
60 template <
typename Po
intT>
bool
62 const Indices &samples, Eigen::VectorXf &model_coefficients)
const
65 if (samples.size () != sample_size_)
67 PCL_ERROR (
"[pcl::SampleConsensusModelSphere::computeModelCoefficients] Invalid set of samples given (%lu)!\n", samples.size ());
72 for (
int i = 0; i < 4; i++)
74 temp (i, 0) = (*input_)[samples[i]].x;
75 temp (i, 1) = (*input_)[samples[i]].y;
76 temp (i, 2) = (*input_)[samples[i]].z;
79 float m11 = temp.determinant ();
85 for (
int i = 0; i < 4; ++i)
87 temp (i, 0) = ((*input_)[samples[i]].x) * ((*input_)[samples[i]].x) +
88 ((*input_)[samples[i]].y) * ((*input_)[samples[i]].y) +
89 ((*input_)[samples[i]].z) * ((*input_)[samples[i]].z);
91 float m12 = temp.determinant ();
93 for (
int i = 0; i < 4; ++i)
95 temp (i, 1) = temp (i, 0);
96 temp (i, 0) = (*input_)[samples[i]].x;
98 float m13 = temp.determinant ();
100 for (
int i = 0; i < 4; ++i)
102 temp (i, 2) = temp (i, 1);
103 temp (i, 1) = (*input_)[samples[i]].y;
105 float m14 = temp.determinant ();
107 for (
int i = 0; i < 4; ++i)
109 temp (i, 0) = temp (i, 2);
110 temp (i, 1) = (*input_)[samples[i]].x;
111 temp (i, 2) = (*input_)[samples[i]].y;
112 temp (i, 3) = (*input_)[samples[i]].z;
114 float m15 = temp.determinant ();
117 model_coefficients.resize (model_size_);
118 model_coefficients[0] = 0.5f * m12 / m11;
119 model_coefficients[1] = 0.5f * m13 / m11;
120 model_coefficients[2] = 0.5f * m14 / m11;
122 model_coefficients[3] = std::sqrt (model_coefficients[0] * model_coefficients[0] +
123 model_coefficients[1] * model_coefficients[1] +
124 model_coefficients[2] * model_coefficients[2] - m15 / m11);
130 template <
typename Po
intT>
void
132 const Eigen::VectorXf &model_coefficients, std::vector<double> &distances)
const
135 if (!isModelValid (model_coefficients))
140 distances.resize (indices_->size ());
143 for (std::size_t i = 0; i < indices_->size (); ++i)
147 distances[i] = std::abs (std::sqrt (
148 ( (*input_)[(*indices_)[i]].x - model_coefficients[0] ) *
149 ( (*input_)[(*indices_)[i]].x - model_coefficients[0] ) +
151 ( (*input_)[(*indices_)[i]].y - model_coefficients[1] ) *
152 ( (*input_)[(*indices_)[i]].y - model_coefficients[1] ) +
154 ( (*input_)[(*indices_)[i]].z - model_coefficients[2] ) *
155 ( (*input_)[(*indices_)[i]].z - model_coefficients[2] )
156 ) - model_coefficients[3]);
161 template <
typename Po
intT>
void
163 const Eigen::VectorXf &model_coefficients,
const double threshold,
Indices &inliers)
166 if (!isModelValid (model_coefficients))
173 error_sqr_dists_.clear ();
174 inliers.reserve (indices_->size ());
175 error_sqr_dists_.reserve (indices_->size ());
178 for (std::size_t i = 0; i < indices_->size (); ++i)
180 double distance = std::abs (std::sqrt (
181 ( (*input_)[(*indices_)[i]].x - model_coefficients[0] ) *
182 ( (*input_)[(*indices_)[i]].x - model_coefficients[0] ) +
184 ( (*input_)[(*indices_)[i]].y - model_coefficients[1] ) *
185 ( (*input_)[(*indices_)[i]].y - model_coefficients[1] ) +
187 ( (*input_)[(*indices_)[i]].z - model_coefficients[2] ) *
188 ( (*input_)[(*indices_)[i]].z - model_coefficients[2] )
189 ) - model_coefficients[3]);
195 inliers.push_back ((*indices_)[i]);
196 error_sqr_dists_.push_back (
static_cast<double> (
distance));
202 template <
typename Po
intT> std::size_t
204 const Eigen::VectorXf &model_coefficients,
const double threshold)
const
207 if (!isModelValid (model_coefficients))
210 std::size_t nr_p = 0;
213 for (std::size_t i = 0; i < indices_->size (); ++i)
217 if (std::abs (std::sqrt (
218 ( (*input_)[(*indices_)[i]].x - model_coefficients[0] ) *
219 ( (*input_)[(*indices_)[i]].x - model_coefficients[0] ) +
221 ( (*input_)[(*indices_)[i]].y - model_coefficients[1] ) *
222 ( (*input_)[(*indices_)[i]].y - model_coefficients[1] ) +
224 ( (*input_)[(*indices_)[i]].z - model_coefficients[2] ) *
225 ( (*input_)[(*indices_)[i]].z - model_coefficients[2] )
226 ) - model_coefficients[3]) < threshold)
233 template <
typename Po
intT>
void
235 const Indices &inliers,
const Eigen::VectorXf &model_coefficients, Eigen::VectorXf &optimized_coefficients)
const
237 optimized_coefficients = model_coefficients;
240 if (!isModelValid (model_coefficients))
242 PCL_ERROR (
"[pcl::SampleConsensusModelSphere::optimizeModelCoefficients] Given model is invalid!\n");
247 if (inliers.size () <= sample_size_)
249 PCL_ERROR (
"[pcl::SampleConsensusModelSphere::optimizeModelCoefficients] Not enough inliers to refine/optimize the model's coefficients (%lu)! Returning the same coefficients.\n", inliers.size ());
253 OptimizationFunctor functor (
this, inliers);
254 Eigen::NumericalDiff<OptimizationFunctor> num_diff (functor);
255 Eigen::LevenbergMarquardt<Eigen::NumericalDiff<OptimizationFunctor>,
float> lm (num_diff);
256 int info = lm.minimize (optimized_coefficients);
259 PCL_DEBUG (
"[pcl::SampleConsensusModelSphere::optimizeModelCoefficients] LM solver finished with exit code %i, having a residual norm of %g. \nInitial solution: %g %g %g %g \nFinal solution: %g %g %g %g\n",
260 info, lm.fvec.norm (), model_coefficients[0], model_coefficients[1], model_coefficients[2], model_coefficients[3], optimized_coefficients[0], optimized_coefficients[1], optimized_coefficients[2], optimized_coefficients[3]);
264 template <
typename Po
intT>
void
266 const Indices &,
const Eigen::VectorXf &model_coefficients,
PointCloud &projected_points,
bool)
const
269 if (!isModelValid (model_coefficients))
271 PCL_ERROR (
"[pcl::SampleConsensusModelSphere::projectPoints] Given model is invalid!\n");
276 projected_points.
points.resize (input_->size ());
277 projected_points.
header = input_->header;
278 projected_points.
width = input_->width;
279 projected_points.
height = input_->height;
280 projected_points.
is_dense = input_->is_dense;
282 PCL_WARN (
"[pcl::SampleConsensusModelSphere::projectPoints] Not implemented yet.\n");
283 projected_points.
points = input_->points;
287 template <
typename Po
intT>
bool
289 const std::set<index_t> &indices,
const Eigen::VectorXf &model_coefficients,
const double threshold)
const
292 if (!isModelValid (model_coefficients))
294 PCL_ERROR (
"[pcl::SampleConsensusModelSphere::doSamplesVerifyModel] Given model is invalid!\n");
298 for (
const auto &index : indices)
303 ( (*input_)[index].x - model_coefficients[0] ) *
304 ( (*input_)[index].x - model_coefficients[0] ) +
305 ( (*input_)[index].y - model_coefficients[1] ) *
306 ( (*input_)[index].y - model_coefficients[1] ) +
307 ( (*input_)[index].z - model_coefficients[2] ) *
308 ( (*input_)[index].z - model_coefficients[2] )
309 ) - model_coefficients[3]) > threshold)
318 #define PCL_INSTANTIATE_SampleConsensusModelSphere(T) template class PCL_EXPORTS pcl::SampleConsensusModelSphere<T>;
PointCloud represents the base class in PCL for storing collections of 3D points.
bool is_dense
True if no points are invalid (e.g., have NaN or Inf values in any of their floating point fields).
std::uint32_t width
The point cloud width (if organized as an image-structure).
pcl::PCLHeader header
The point cloud header.
std::uint32_t height
The point cloud height (if organized as an image-structure).
std::vector< PointT, Eigen::aligned_allocator< PointT > > points
The point data.
bool isSampleGood(const Indices &samples) const override
Check if a sample of indices results in a good sample of points indices.
void getDistancesToModel(const Eigen::VectorXf &model_coefficients, std::vector< double > &distances) const override
Compute all distances from the cloud data to a given sphere model.
void optimizeModelCoefficients(const Indices &inliers, const Eigen::VectorXf &model_coefficients, Eigen::VectorXf &optimized_coefficients) const override
Recompute the sphere coefficients using the given inlier set and return them to the user.
std::size_t countWithinDistance(const Eigen::VectorXf &model_coefficients, const double threshold) const override
Count all the points which respect the given model coefficients as inliers.
bool doSamplesVerifyModel(const std::set< index_t > &indices, const Eigen::VectorXf &model_coefficients, const double threshold) const override
Verify whether a subset of indices verifies the given sphere model coefficients.
void projectPoints(const Indices &inliers, const Eigen::VectorXf &model_coefficients, PointCloud &projected_points, bool copy_data_fields=true) const override
Create a new point cloud with inliers projected onto the sphere model.
bool computeModelCoefficients(const Indices &samples, Eigen::VectorXf &model_coefficients) const override
Check whether the given index samples can form a valid sphere model, compute the model coefficients f...
void selectWithinDistance(const Eigen::VectorXf &model_coefficients, const double threshold, Indices &inliers) override
Select all the points which respect the given model coefficients as inliers.
float distance(const PointT &p1, const PointT &p2)
IndicesAllocator<> Indices
Type used for indices in PCL.