A vehicle-mounted point cloud clustering method and system

Abstract

The present invention proposes a vehicle-mounted point cloud clustering method and system, including point cloud denoising and filtering, including removing scattered noise points from the point cloud, and performing point cloud filtering processing to distinguish ground points and non-ground points; The cloud is segmented to generate supervoxels, including non-ground points, which are segmented using a density-based spatial clustering algorithm to generate supervoxels to ensure that there is no adhesion between different ground objects; spatial context associated point cloud clustering, including analyzing supervoxels The characteristics of voxels and spatial context are associated, and multi-factor weights are integrated to perform super-voxel region growth and complete point cloud clustering. The invention improves the problem of over-segmentation or under-segmentation of vehicle-mounted point cloud clustering, satisfies the demand for fast acquisition of three-dimensional space information, and has important market value.

Description

technical field [0001] The invention relates to the technical field of vehicle-mounted laser scanning data processing, in particular to a vehicle-mounted point cloud clustering method and system combined with spatial context association. Background technique [0002] The point cloud data acquired by the vehicle-borne laser scanning system (Vehicle-borne Laser Scanning, VLS) has the characteristics of high density, high precision, and fast acquisition. At present, scholars at home and abroad have conducted a lot of research on the application of point cloud data, among which vehicle-mounted point cloud clustering is an important part of VLS data processing and information extraction, and is the premise and key link to realize automatic recognition of ground objects. [0003] The current common point cloud clustering methods can be divided into partition clustering, hierarchical clustering, grid clustering and density clustering. The division clustering method depends on the ...

AI Technical Summary

Problems solved by technology

However, urban streetscapes are densely populated with vertically distributed features, and the adhesion between features is serious. In this case, density clustering that only considers the spatial distance between points will cause serious under-segmentation or over-segmentation. Happening

Images