Article Directory
1. Introduction
As a prerequisite for many applications, point cloud segmentation is directly related to the effect of subsequent processing such as surface reconstruction and feature extraction using point cloud data. As a more classic clustering and segmentation algorithm, the region growing algorithm has a wide range of applications. The algorithm process is as follows:
1. First, sort the curvature values of each point from small to large.
2. Select the point with the smallest curvature value as the initial seed point, add it to the seed point set P, and start the entire growth process. Why choose the point of minimum curvature? Mainly because, in general, locations with smaller curvature values are flatter, and growing from the flattest regions reduces the number of final segmented point clouds.
3. Traverse each seed point in the seed point set to find the seed point P i P_i