Euclidean Clustering Segmentation With Normals

Description

This operator performs clustering segmentation on the input point cloud based on the Euclidean distance between points. It aggregates spatially close points into independent clusters. If the distance between a point and any point in an existing cluster is less than the set "Search Radius", or if it can connect to a point within the radius, the point is assigned to that cluster. The operator finally outputs a list of point clouds, where each point cloud represents a segmented cluster.

Overview

Uses a connected component algorithm based on Euclidean distance to group points with spatial distance below a threshold into the same cluster, suitable for scenes where objects have obvious spatial separation.

Scenarios

  • Instance Segmentation: Segmenting physically separated objects in the scene into independent point cloud clusters. For example, segmenting multiple parts on a conveyor belt or multiple cups on a table.

  • Noise Removal: By setting "Minimum Cluster Points", point cloud clusters with too few points, which may be background noise or sensor errors, can be effectively filtered out.

  • Point Cloud Preprocessing: Segmenting the raw point cloud or point cloud within an ROI into units representing different objects before performing tasks like object recognition or pose estimation.

  • Functionally very similar to the "Connected Component Segmentation" operator, as both usually utilize spatial proximity for clustering.

Inputs and Outputs

Inputs

Point Cloud: The input point cloud or list of point clouds to be segmented.

Outputs

Segmented point cloud: The list of point cloud clusters obtained after segmentation.

Parameters

Search Radius

Description

  • Decrease this value: Makes the clustering conditions stricter. Only very close points will be grouped into the same cluster, which may cause a single object to be split into multiple small clusters, or make it easier to separate objects that are very close.

  • Increase this value: Makes the clustering conditions looser. Points that are slightly further apart may be grouped into one cluster, potentially merging different objects.

Parameter Range

[0, 200], Default value: 5, Unit: mm

Parameter Range

Values/Range: [0, 200]

Default: 5

Minimum Cluster Points

Description

Used to filter out point cloud clusters that are too small. Increasing this value can remove more small clusters (usually considered noise), but if set too high, it may filter out valid objects with smaller sizes.

Parameter Range

[1, 10000000], Default value: 100

Parameter Range

Values/Range: [1, 10000000]

Default: 1000

Maximum Cluster Points

Description

Used to filter out point cloud clusters that are too large. Suitable for scenarios where extra-large clustering results such as background or ground need to be excluded. Usually, keep the default value (a very large number) unless there is a special requirement.

Parameter Range

[1, 10000000], Default value: 4000000

Parameter Range

Values/Range: [1, 10000000]

Default: 4000000

Keep All Results

Description

Determines whether to output all clustering results that meet the size conditions.

Tuning Instructions

  • Enabled (Default): Retains all clusters filtered by the minimum/maximum number of points.

  • Disabled: Retains only the top N clusters with the most points, where N is specified by the "Number of Results to Keep" parameter.

Parameter Range

Values/Range:

  • True

  • False

Default: True

Number of Results to Keep

Description

The operator sorts all segmentation results by point cloud quantity from large to small. If this parameter is set to N, only the top N sorted point cloud clusters are output. For example, setting it to 1 means only outputting the cluster with the most points.

Parameter Range

[1, 1000], Default value: 1

Parameter Range

Values/Range: [1, 1000]

Default: 1

Enable Sorting

Description

Determines whether to sort the final output list of point cloud clusters by point count from large to small.

Tuning Instructions

  • Enabled (Default): The output point cloud list is sorted from most points to fewest.

  • Disabled: The order of the output point cloud list is indeterminate.

If subsequent processing depends on the size order of point cloud clusters, this item should be kept enabled.

Parameter Range

Default: True

Enable Node

Description

Controls whether this operator executes its computation.

Tuning Instructions

  • Enabled (Default): Runs the operator function normally.

  • Disabled: The operator performs no operation and directly outputs the input data.

Parameter Range

Default: True