Method and device for extracting plane features of multi-line LiDAR point cloud data

Abstract

The invention relates to a method and device for extracting plane features of multi-line LiDAR point cloud data. The method comprises the following steps: acquiring original point cloud data of multi-line LiDAR to obtain a distance image comprising N scanning lines; finding out a breakpoint of each scanning line, and dividing each scanning line into a plurality of sub-segments by utilizing breakpoints; growing on scanning lines and among the scanning lines by taking sub-segments as units to obtain merged sub-segments; determining voting scores of the combined sub-segments on each potential plane according to the areas of the combined sub-segments, the dimension feature descriptors, the direction with the minimum point set change and the normal vectors of the voted planes, and selecting a plurality of potential planes with higher scores as candidate planes of the combined sub-segments; selecting the combined sub-segments voted for the candidate planes to perform plane fitting to obtaina plurality of plane patches; and carrying out region growth by taking points as units on the plurality of plane sheets, and calculating parameters of each plane sheet after growth is stopped, therebycompleting extraction of plane features. The plane features of the multi-line LiDAR point cloud can be accurately and completely extracted.

Description

technical field [0001] The invention relates to a method and device for extracting plane features of multi-line LiDAR point cloud data, belonging to the technical field of point cloud data processing. Background technique [0002] As an active sensor, LiDAR (Light Detection and Ranging, LiDAR) usually uses invisible light in the 800-1000nm band as the light source. The light in this band will not interfere with the ambient light, so it can work under different lighting conditions. . The distance measurement of LiDAR is based on the time-of-flight or phase difference principle, which can output the return light intensity information. According to the measurement dimension, LiDAR can be divided into 2D LiDAR and 3D LiDAR. There are three types of 3D LiDAR, 3D LiDAR customized for rotation-based 2D LiDAR (also called Ro-LiDAR), Terrestrial Laser Scanner (TLS) and multi-line LiDAR. Among them, TLS has a measurement accuracy of millimeter level and is expensive, and is mainly ...

AI Technical Summary

Problems solved by technology

[0005] However, the multi-line LiDAR point cloud has a strong sparsity. The existing plane extraction methods are mainly aimed at relatively dense point clouds. It is often difficult to directly apply these extraction methods to the relatively sparse multi-line LiDAR point cloud. satisfactory results, such as figure 2 As shown, the point cloud distribution is extremely sparse in some positions or directions, which may lead to poor plane extraction effect, and may have over-segmentation or under-segmentation problems. At the same time, sparse areas are often missed or falsely detected. For this reason, It is necessary to propose a technical solution for multi-line LiDAR point cloud plane feature extraction

Images