Classification method based on vehicle-mounted LiDAR point cloud data

Abstract

The invention discloses a classification method based on vehicle-mounted LiDAR point cloud data. The method comprises the steps that streetscape vehicle-mounted LiDAR point cloud data are obtained and preprocessed so as to remove redundant data and noise in the point cloud data, and the point cloud normal vector, curvature and density are calculated in the preprocessing process; the preprocessed point cloud data are classified; according to the point cloud normal vector, curvature and density, power lines, building vertical face points and tree points are extracted. According to the classification method, implementation is easy, calculated quantity is small, implementation is fast and effective, the foundation is laid for streetscape facet extraction, and facet extraction is more accurate.

Description

technical field [0001] The invention relates to the technical field of vehicle-mounted LiDAR point cloud data processing, in particular to a classification method based on vehicle-mounted LiDAR point cloud data that can effectively extract building points, power lines and tree points. Background technique [0002] With the digitalization of cities and the increasing demand for information technology, street view maps are developing more and more rapidly. As an advanced measurement method, vehicle-mounted laser scanning not only has the characteristics of fast, non-contact, real-time, dynamic, active, high-density and high-precision, but also can collect large-area three-dimensional space data and acquire buildings, roads, etc. The surface information of urban features such as vegetation and vegetation provides ideas for the extraction of street scene patches. Therefore, how to quickly and accurately classify vehicle-mounted laser point cloud data has become one of the first...

AI Technical Summary

Problems solved by technology

[0004] Foreign countries have developed a relatively ideal automatic classification and recognition algorithm for point cloud data in this field: such as an algorithm for classifying laser scanning point cloud data into different groups according to the point spatial distribution characteristics (geometric features, dispersion degree and density information) of cross-sectional scanning points , but the algorithm’s recognition of mixed arrangement points is relatively complicated; the point cloud data feature extraction method based on building semantics aims to construct walls, doors, windows, and building protrusions by extracting semantic features from ground point cloud data and recessed parts, as well as the digital model of parts such as the roof, the processing process is also more complicated

Images