On-board LiDAR point cloud road information extraction method and device

Abstract

The invention provides an on-board LiDAR point cloud road information extraction method and device. The method comprises the following steps that point cloud data of a to-be-detected road is filteredto obtain ground point cloud data; a two-dimensional net is established for ground point cloud data; first average elevation value for each grid, a highest elevation, a lowest elevation for and secondaverage elevation value of all adjacent grids are determined, existence of road point grids in all the adjacent grids is determined according to values; when the road point grids exist, grid clustering is carried out for all road point grids to determine road point grid clusters; and according to the size of each road point grid cluster, whether the point cloud data in the road point grid clusterare road point cloud data is determined. According to the technical scheme, a standard spatial grid is established for the point cloud, location of the road is determined by the spatial distributioncharacteristics and the relationship between elevation differences; and the grids with roads are determined according to the grid cluster growth mode to achieve the aim of road extraction.

Description

technical field [0001] The invention relates to the technical field of laser point cloud classification, in particular to an airborne LiDAR point cloud road information extraction method and device. Background technique [0002] Airborne Lidar (LIDAR) technology is a technology developed in recent years that can directly obtain 3D information on the ground. It has incomparable advantages over traditional aerial photogrammetry and has been widely used in urban building extraction and 3D reconstruction. However, the point cloud data obtained by LIDAR contains not only building information, but also other non-building information, such as ground information, vegetation information, and road information. It is usually difficult to segment roads and surrounding ground points using height information, which makes the automatic extraction of road information more difficult than that of buildings. There are relatively few research results, and road information is getting more and mo...

AI Technical Summary

Problems solved by technology

[0005] The embodiment of the present invention provides an airborne LiDAR point cloud road information extraction method and device, using reflection intensity information for extraction, and using filtering to eliminate interference from non-ground points, reducing calculation time, and solving the effect of reflection intensity information extraction Due to instability and low efficiency, the actual production application does not achieve good results, and the rapid extraction of standard roads under complex terrain is realized.

Images