Street tree target recognition method based on vehicle-mounted 2d LiDAR point cloud data

Abstract

The invention discloses a street tree target recognition method based on vehicle-mounted 2D LiDAR point cloud data, which includes using the vehicle-mounted 2D LiDAR system to obtain street data, saving it as a point cloud file, marking the point cloud file with tree crowns and non-tree crown categories; constructing a variable scale Grid, extract the three-dimensional spherical domain from the variable-scale grid; calculate the point cloud feature vector in the three-dimensional spherical domain, and obtain the point cloud feature vector set; the SVM algorithm learns the tree crown point cloud classifier from the point cloud feature vector set; according to the tree crown point cloud The classifier performs online recognition on the point cloud frame to obtain the spray prescription map. The invention adopts the vehicle-mounted 2D LiDAR with small data redundancy and easy online processing to acquire urban street data, and provides accurate spray basis for target application. Establish a point cloud sequence spatial index structure that preserves data accuracy, improves neighborhood retrieval efficiency, and meets online processing requirements to achieve accurate identification of street tree targets.

Description

technical field [0001] The invention relates to the field of target identification, and more particularly, to a method for identifying a street tree target based on vehicle-mounted 2D LiDAR point cloud data. Background technique [0002] Street trees are an important part of urban ecosystems and urban landscapes. Facing the current increasingly severe environmental pollution problems, street trees are of great significance in improving the urban ecological environment, purifying the air, regulating the climate and conserving water sources. Affected by the greenhouse effect, human interference and other factors, the diseases and insect pests of street trees are increasing day by day, resulting in the withering or death of street trees, which not only seriously affects the greening and beautification effect of street trees, but also directly affects the urban ecological environment and residents' life, which has become a restrictive factor for the refined management of landscap...

AI Technical Summary

Problems solved by technology

Uniform spatial index structures such as grids and voxels can effectively reduce the amount of point cloud data and data redundancy, and improve the efficiency of neighborhood retrieval, but it is difficult to deal with uneven point cloud densities. At the same time, grid / voxel size selection and segmentation Accuracy is difficult to balance

Spatial index structures such as quadtrees, octrees, and KD trees can better adapt to the uneven distribution of data, retain data accuracy, and facilitate batch display and storage of point clouds, but they are not suitable for online data processing.

Images