Point Cloud Axial Merge

Description

This operator is used to merge multiple input point cloud clusters along a specified axis. It first calculates the center and size of each point cloud cluster, then sorts them along the specified "Merge Axis". It then iterates through the sorted point cloud clusters; if two adjacent clusters are close enough along the axis and sufficiently similar in size, they are merged into a larger cluster. This process continues until no more clusters can be merged.

Overview

Merges multiple input point cloud clusters along a specified axis, used to merge multiple fragments of the same object that have been over-segmented.

Scenarios

Merging multiple independent parts belonging to a group that are arranged along an axis into a whole. For example, merging all items detected on the same shelf layer into a "layer" entity.

Inputs and Outputs

Inputs

Scene Point Cloud: The list of point clouds to be merged.

Outputs

Merged point cloud: Returns the list of point cloud clusters merged according to the specified axis and parameters.

Parameters

Axial Distance Threshold

Description

  • Increase this value: Relaxes the merge condition, allowing clusters further apart to be merged.

  • Decrease this value: Requires clusters to be very close to merge.

Set this value according to the actual gap size between parts of the object.

Parameter Range

[1, 5000], Default value: 50, Unit: mm

Parameter Range

Values/Range: [1, 5000]

Default: 50

Unit: mm

Axial Size Ratio Threshold

Description

  • Set to 0: Disables this size check completely; merging is determined solely by the "Axial Distance Threshold".

  • Set close to 1 (e.g., 0.8-0.9): Requires the sizes of the two clusters to be very close to merge.

  • Set to a smaller value (e.g., 0.1-0.3): Allows clusters with large size differences to merge.

This parameter helps prevent a large object from being incorrectly merged with a small fragment next to it.

Parameter Range

[0, 1], Default value: 0.1

Parameter Range

Values/Range: [0, 1]

Default: 0.1

Merge Axis

Description

Defines a 3D vector [x, y, z] as the directional reference for distance and size measurement.

Tuning Instructions

Sets the target direction for the merge operation.

Parameter Range

Default: [0, 0, 1]

Minimum Cluster Points

Description

Suitable for filtering out meaningless small noise clusters composed of a few points. Appropriately increase this value to ensure meaningful clustering results are output.

Parameter Range

[1, 4000000], Default value: 100

Parameter Range

Values/Range: [1, 10000000]

Default: 1000

Maximum Cluster Points

Description

Suitable for scenarios where a very large background or support surface needs to be excluded. For example, to ignore a table and only analyze small objects on it, set an upper limit to filter out the largest point cloud cluster representing the table.

Parameter Range

[1, 4000000], Default value: 4000000

Parameter Range

Values/Range: [1, 10000000]

Default: 4000000

Sort Order

Description

Sets the sorting rule for the output list of point cloud clusters.

Tuning Instructions

The operator calculates the position of the center point of each cluster on the "Clustering Axis" and sorts them based on this position.

"Low to High" means sorting from the negative direction to the positive direction along the clustering axis, and vice versa for sorting from the positive direction to the negative direction.

Parameter Range

Values/Range:

  • Low to High

  • High to Low

Default: Low to High

Keep All Results

Description

Controls whether to output all segmentation results or only retain a portion for each input point cloud.

Tuning Instructions

Enabled (True): Default option, retains all clusters found that meet the point count limits. Disabled (False): Will only retain the top N largest clusters specified by the "Number of Results to Keep" parameter.

Parameter Range

Values/Range:

  • True

  • False

Default: True

Number of Results to Keep

Description

For each input point cloud, sorts the segmentation results by point cloud quantity from large to small, retaining the top few point clouds. For example, setting to 1 means retaining only the largest result, 2 means retaining the top 2 results, 3 means retaining the top 3 results, and so on.

Parameter Range

[1, 1000], Default value: 1

Parameter Range

Values/Range: [1, 100]

Default: 1

Enable Node

Description

Controls whether the operator performs the operation.

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