Filtering method for laser radar point cloud trailing point

Abstract

The invention provides a filtering method for multi-line laser radar point cloud tailing points. The method comprises the following steps: firstly, filtering the trailing points of the point clouds generated by each line laser; secondly, for the point cloud generated by each line laser, calculating the distance between each point and the left and right adjacent points, and if the two distance values are both greater than a set distance threshold value, preliminarily judging that the point is a trailing point; for the point which is preliminarily judged to be the trailing point, calculating a left adjacent point of the point, an angle corresponding to the point cloud original point in a triangle formed by the point and the point cloud original point, and a right adjacent point of the point, an angle corresponding to the point cloud original point in a triangle formed by the point and the point cloud original point, and if the angle values of the two angles are both smaller than a set angle threshold value, considering that the point is a trailing point and filtered; and finally, superposing the point clouds generated by each line laser of which the trailing points are filtered to generate a frame of point clouds of which the trailing points are filtered. The method has the advantages of being low in calculation amount and simple and effective in algorithm implementation.

Description

technical field [0001] The invention relates to the fields of laser radar scanning mapping and perception, in particular to a method for filtering trailing points in a multi-line laser radar point cloud. Background technique [0002] With the rapid development of laser technology and computer technology, lidar measurement has become a new technology to efficiently obtain high-precision and reliable three-dimensional data. [0003] This technology can quickly, accurately, and contactlessly acquire 3D point cloud information on the surface of complex objects, and complete the 3D reconstruction of entities. It has been widely used in various fields such as geographic information systems, robot navigation, and automatic driving. [0004] However, due to the influence of the physical characteristics of the scanning equipment and the characteristics of the laser, the obtained original point cloud data will have trailing points. Using the original point cloud data containing traili...

AI Technical Summary

Problems solved by technology

[0004] However, due to the influence of the physical characteristics of the scanning equipment and the laser characteristics, the obtained original point cloud data will have trailing points. Using the original point cloud data containing trailing points for 3D reconstruction directly will cause a large number of irregular noise points, resulting in It is difficult to remove, so it is necessary to preprocess each frame of point cloud to filter out trailing points, but the existing methods cannot achieve a good filtering effect

Images