Point cloud data partitioning method based on three-dimensional laser radar

Abstract

The invention relates to a point cloud data partitioning method based on three-dimensional laser radar. The method comprises the following steps of: (1) establishing a radar coordinate system oxyz of tested vehicles; (2) preprocessing the radar data acquired by the laser radar, establishing a region of interest under the radar coordinate system oxyz and filtering out ground noise; (3) establishing an image coordinate system o'uv, and defining the mapping relation between the radar coordinate system oxyz and the image coordinate system o'uv; (4) testing vehicles on a radar cloud picture directly by using an image processing algorithm, and characterizing all tested vehicles (except the testing vehicles) in the region of interest by using a bounding box so as to obtain four vertex coordinates of the bounding box of the tested vehicles under the image coordinate system o'uv; and (5) mapping the four vertex coordinates of the bounding box under the image coordinate system o'uv in step (4) to the radar coordinate system oxyz according to the coordinate mapping model so as to obtain the shape vector parameters of the tested vehicles according to the position vector parameter of the tested vehicles. The method is high in operation efficiency, test precision and reliability, and can be widely applied to the technical field of vehicle environment sensing.

Description

technical field [0001] The invention relates to the technical field of environment perception of modern traffic intelligent vehicles, in particular to a point cloud data segmentation method based on three-dimensional laser radar used in vehicle detection. Background technique [0002] Three-dimensional lidar is one of the important sensors for intelligent vehicles to obtain external information. It has the advantages of high reliability, real-time performance and high accuracy, so it is widely used in the research of intelligent vehicle environment perception. The radar has multiple laser sensors, and each sensor measures discrete data points which can be denoted as P i (x, y, z, s), where x, y, and z represent the three-dimensional physical distance in meters, and s represents the reflection intensity, which is a dimensionless value. The point cloud data is a collection of large-scale discrete measurement point data{P 1 ,P 2 ,P 3 ,...,P N}, collecting about 400,000 mea...

AI Technical Summary

Problems solved by technology

However, this method is more complicated, which is not conducive to engineering realization, and at the same time, it cannot handle static and low-speed vehicles well.

Images