Hierarchical airborne LiDAR point cloud classification method using geometric and intensity characteristics

Abstract

The invention discloses a hierarchical airborne LiDAR point cloud classification method using geometric and intensity characteristics, and belongs to the technical field of remote sensing science. Themethod comprises: firstly, acquiring three-dimensional geometrical information and intensity information of the ground surface through airborne LiDAR, and constructing geometrical characteristics andintensity characteristics for each LiDAR point according to the geometrical information and the intensity information; processing the geometric features by using a random forest classifier to obtaina supervised classification result; extracting ground features from the supervised classification result, and processing the strength characteristics of the ground feature points by using a Gaussian mixture model to obtain an unsupervised classification result; and fusing the supervised classification result and the unsupervised classification result by using a heuristic rule to obtain a final classification result. According to the airborne LiDAR point cloud classification method, the defects that an airborne LiDAR point cloud supervision classifier is unstable and difficult to migrate due toeasily-changed intensity information in the prior art are overcome, and the geometric information and the intensity information of the airborne LiDAR point cloud can be utilized in a layered mode toobtain a good airborne LiDAR point cloud classification result.

Description

technical field [0001] The invention relates to the field of remote sensing science and technology, more specifically, to a hierarchical airborne LiDAR point cloud classification method using geometric and intensity features. Background technique [0002] LiDAR refers to Light Detection And Ranging (LiDAR for short), that is, laser radar. LiDAR uses the Global Positioning System (GPS for short) and the Inertial Measurement Unit (IMU for short) to directly obtain the three-dimensional geometric information of surface objects. The data measured by LiDAR is represented by discrete points, so LiDAR data is also called point cloud data; applying classification technology to interpret objects such as buildings, vegetation, cars, and ground in these original point clouds is an important aspect of urban form and ecological research. an indispensable part of the process. However, different types of ground features, such as buildings, vegetation, cars, ground, etc., often appear in ...

AI Technical Summary

Problems solved by technology

[0005] However, due to the influence of flight altitude, transmission power, atmospheric transmittance, etc., the intensity information recorded by airborne LiDAR is variable, resulting in the classifier obtained by direct training of supervised learning using geometric information and intensity information. Poor, difficult to migrate to airborne LiDAR data in other regions

Images