PCLIterativeClosestPoint

MLModule

genre	PCLRegistration
author	Wolf Spindler
package	FMEstable/PCL
dll	MLPCLRegistration
definition	MLPCLRegistration.def
see also	PCLSampleConsensus
inherits from	PCLModule
keywords	match, registration

Purpose

Uses the PCL Iterative Closest Point algorithm to find a transformation from the point cloud at inputPCLObject0 to inputPCLObject1 such that both point clouds match as good as possible.

Windows

Default Panel

Input Fields

inputPCLObject0

name: inputPCLObject0, type: MLBase

Connect the source point cloud which shall be transformed and mapped onto the point cloud at inputPCLObject1 by the calculated transformation after registration.

inputPCLObject1

name: inputPCLObject1, type: MLBase

Connect the so-called target point cloud onto which the point cloud at inputPCLObject0 shall be transformed by the calculated registration. Must contain at least one point to allow valid calculations of the module.

Output Fields

outputPCLObject0

name: outputPCLObject0, type: MLBase

The point cloud resulting from the inputPCLObject0 transformed with the calculation result. Indices created by the algorithm are also provided in the output base object.

Parameter Fields

Field Index

Euclidean Fitness Epsilon: Double	Use Reciprocal Correspondences: Bool
Final Transformation: Matrix
Fitness Score: Double
Last Incremental Transformation: Matrix
Max Correspondence Distance: Double
Maximum Iterations: Integer
Status: String
Transformation Epsilon: Double

Visible Fields

Status

name: status, type: String, persistent: no

Shows status information about processing results, or in case of errors, some information about reasons.

see also PCLModule.status

Maximum Iterations

name: maximumIterations, type: Integer, default: 50

The maximum number of iterations the internal optimization should run for.

Max Correspondence Distance

name: maxCorrespondenceDistance, type: Double, default: 1.34078079299426e+154

The maximum distance threshold between two correspondent points in source and target. If the distance is larger than this threshold, the points will be ignored in the alignment process.

Transformation Epsilon

name: transformationEpsilon, type: Double, default: 1e-08

The transformation epsilon (maximum allowable difference between two consecutive transformations) in order for an optimization to be considered as having converged to the final solution.

Euclidean Fitness Epsilon

name: euclideanFitnessEpsilon, type: Double, default: -1.79769313486232e+308

The maximum allowed Euclidean error between two consecutive steps in the ICP loop, before the algorithm is considered to have converged; the error is estimated as the sum of the differences between correspondences in an Euclidean sense, divided by the number of correspondences.

Final Transformation

name: finalTransformation, type: Matrix, persistent: no

Output only: the final transformation matrix estimated by the registration method.

Last Incremental Transformation

name: lastIncrementalTransformation, type: Matrix, persistent: no

Output only: the last incremental transformation matrix estimated by the registration method.

Fitness Score

name: fitnessScore, type: Double, persistent: no

Output only: the Euclidean fitness score (e.g., sum of squared distances from the source to the target).

Use Reciprocal Correspondences

name: useReciprocalCorrespondences, type: Bool, default: FALSE

Determines whether to use reciprocal correspondence or not.