PCLVoxelGrid

MLModule

genre	PCLFilters
author	Wolf Spindler
package	FMEstable/PCL
dll	MLPCLFilters
definition	MLPCLFilters.def
inherits from	PCLModule
keywords	points, clouds, PCL, sample, align, grid, PointCloudToMLImage

Purpose

Applies the pcl::VoxelGrid filter to the input point cloud and provides the filter result at the two outputs. See pcl::VoxelGrid for details.

This filter is especially useful to sample or downsample a point cloud into an ML image, in a certain sense it can be considered as the inverse operation PCLMLImageToPointCloud, however it operates completely different.

Windows

Default Panel

Input Fields

inputPCLObject0

name: inputPCLObject0, type: MLBase

Connect the point cloud which shall be downsampled and voxelized into the output image.

Output Fields

output0

name: output0, type: Image

The output image generated from the input cloud. It will be invalid if no input point cloud is connected or the sampling result does not contain output points. Otherwise it will have voxel type float and voxel values determined from intensity replacement values (the entry of each point in the cloud which is comparable to an intensity such as label, curvature, intensity etc.) of the points determined as the centroid indexes in the pcl::VoxelGrid filter.

outputPCLObject0

name: outputPCLObject0, type: MLBase

The filter result generated by the pcl::VoxelGrid filter with the input point cloud as source. This module still does not set up any indices in the output base object (which, however, may be added in the future).

Parameter Fields

Field Index

Downsample All Data: Bool	Status: String
Fill Value: Float
Filter Field Name: String
Filter Limits Negative: Bool
Leaf Size: Vector3
Limit Max: Float
Limit Min: Float
Minimum Points Number Per Voxel: Integer

Visible Fields

Status

name: status, type: String, persistent: no

see PCLModule.status

Filter Field Name

name: filterFieldName, type: String, default: x

The name of the point field member to be filtered through the interval, such as “x”, “y”, “z”, “intensity”, “normal_x”, “normal_y”, “normal_z”, “curvature”, “label”, “rgba” etc. If an unknown field name is set then no points are passed through.

Downsample All Data

name: downsampleAllData, type: Bool, default: TRUE

Set to true if all fields need to be downsampled, or false if just XYZ. Disabling this will skip all intensity, curvature, label, normal etc. information from the input cloud.

Limit Min

name: limitMin, type: Float, default: 1

The minimum limit of the filter interval.

Limit Max

name: limitMax, type: Float, default: 0

The maximum limit of the filter interval.

Filter Limits Negative

name: filterLimitsNegative, type: Bool, default: TRUE

Inverts the meaning of the filter interval.

Leaf Size

name: leafSize, type: Vector3, default: 1 1 1

The extent of a leaf, respectively the voxel size of the result image or the size of the cells which shall accumulate points.

Minimum Points Number Per Voxel

name: minimumPointsNumberPerVoxel, type: Integer, default: 0

Set the minimum number of points required in a cell or voxel to gain a valid result voxel or point in the output point cloud.

Fill Value

name: fillValue, type: Float, default: 0

The value to be used for voxels which could not be sampled from the input point cloud.