A laser radar point cloud power line classification method based on normal random sampling distribution

Abstract

The invention discloses a laser radar point cloud power line classification method based on normal random sampling distribution, and the method comprises the following steps: (1) carrying out the preprocessing of original point cloud data, building a digital terrain model, and carrying out the rough extraction of power line candidate points through employing elevation filtering; (2) for the roughly extracted power line candidate point data, further optimizing and extracting power line candidate points in a three-dimensional space based on a normal distribution transformation algorithm; and (3)aiming at the optimized and extracted power line candidate point data, realizing accurate extraction of the power line points in a two-dimensional space by adopting a random sampling consistency algorithm principle. With the method of the invention adopted, power line classification can be realized in laser radar point cloud data of various complex environments such as an urban forest region; andan accurate power line extraction result is provided, so that the point cloud data classification efficiency is greatly improved, a new idea is provided for classification of various point cloud data, and accurate and comprehensive analysis data is provided for work such as power line inspection and the like.

Description

Technical field [0001] The present invention relates to the technical field of data processing, in particular to a method for classifying power lines of lidar point clouds based on normal random sampling distribution. Background technique [0002] As a novel and efficient space detection method, lidar technology can quickly acquire a large number of point cloud data of the target scene with precise three-dimensional space coordinates in a short time, but compared with the progress made in the hardware performance and indicators of the lidar system The software processing of lidar point cloud data is still in its infancy. Facing the massive point cloud data acquired by the lidar hardware system, how to effectively use it is a major issue facing the current lidar point cloud data processing field. problem. At the same time, with the rapid development of my country's economy in various fields, the demand for electricity in all walks of life has grown very rapidly. In the face of l...

AI Technical Summary

Problems solved by technology

[0002] As a novel and efficient means of space detection, lidar technology can quickly acquire a large amount of point cloud data with precise three-dimensional spatial coordinates of the target scene in a short period of time. However, compared with the progress made in hardware performance and indicators of lidar systems , the software processing of LiDAR point cloud data is still in its infancy. Faced with the massive point cloud data acquired by the LiDAR hardware system, how to effectively use it is a major problem in the field of LiDAR point cloud data processing. question

Images