Figure 1.
Diagram of point cloud segmentation.
Figure 2.
Diagram of watershed algorithm
Figure 3.
Regional growth diagram
Figure 4.
Diagram of point cloud region growth
Figure 5.
Clustering algorithm flow diagram
Figure 6.
Undirected graph
TABLE I.
COMPARISON OF VARIOUS SEGMENTATION ALGORITHMS
| Segmentation type | Tightness | To deal with noise | The processing time | Application Scenarios |
|---|---|---|---|---|
| Edge | worst | difficult | fast | Large scene segmentation with a priori capability |
| Region | best | perfect | Depends on the number of point | High precision segmentation scene |
| Cluster | better | suitable for detecting some objects with special shape from the chaotic point cloud | fast | Clustering is the division of a data set into different classes according to a particular criterion |
| Graph | worst | general | It depends on the number of edges | More for image segmentation |