A method for classifying ground points from a thick laser point cloud carried by an unmanned aerial vehicle (UAV).

CN115272889BActive Publication Date: 2026-05-26NANJING RES INST OF SURV MAP & GEOTECH INVESTIG CO LTD

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
NANJING RES INST OF SURV MAP & GEOTECH INVESTIG CO LTD
Filing Date
2022-07-25
Publication Date
2026-05-26

AI Technical Summary

Technical Problem

Existing UAV-borne low-end laser scanning equipment collects point clouds with large thickness and low accuracy, which is difficult to meet the requirements of road completion surveying and high-precision topographic surveying.

Method used

A ground point classification method for thick laser point clouds carried by UAVs is proposed, which includes data acquisition and analysis, extraction of intermediate points in the point cloud, surface fitting, and classification steps. By correcting layering and misalignment phenomena, selecting approximate planar point clouds, calculating the centroid, fitting the surface, and using classification algorithms, the method extracts and classifies the ground point cloud.

Benefits of technology

It improves the accuracy and applicability of low-end laser scanning equipment, enabling it to meet the needs of road completion surveying and high-precision topographic surveying, saving equipment costs and improving production efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115272889B_ABST
    Figure CN115272889B_ABST
Patent Text Reader

Abstract

This invention discloses a method for classifying ground points from a thick UAV-borne LiDAR point cloud, comprising the following steps: First, acquiring and analyzing UAV-borne LiDAR point cloud data; second, extracting points from the middle portion of the thick point cloud and identifying the "middle points"; third, automatically classifying the thick point cloud into ground points based on the "middle points". This classification provides a data foundation for subsequent refined classification, enabling it to be matched with data collected by higher-level equipment.
Need to check novelty before this filing date? Find Prior Art