A point cloud segmentation method based on vector estimation and color clustering

For automatic processing of point clouds, the segmentation is a key but difficult step. Many researchers have tried to develop segmentation methods including edge-based segmentation, surface-based segmentation and color-based segmentation, and so on. In this paper, we present a point data segmentation method based on normal vector estimation and color clustering. The main workflow of this method is made by calculating point normal vector, transforming vector into color, clustering color point and segmenting the raw points set at last. The proposed method combined the advantage of geo-metrical segmentation and color-metrical segmentation. It has been applied to LiDAR point data obtained by ALS (airborne laser scanner), the experiment result show that the segmentation method is promising.