基于移动激光扫描点云数据的行道树树干提取方法

一种基于移动激光扫描点云数据的行道树树干提取方法，它包括以下步骤：构建城市街道点云标注数据集，提取深度、高程、维度、密度、强度等14个局部点云特征；采用boosting监督学习算法自动学习树干点云与非树干点云的差异性，通过特征融合获取高精度树干点云检测器；基于树干点云识别结果，分割标识出每一根树干。本发明的方法能准确提取行道树树干，应用于行道树资源调查领域。

A trunk extraction method for street trees based on mobile LiDAR point cloud data includes the following steps: First, construct a labeled urban street point cloud dataset and extract 14 local point cloud features including depth, elevation, dimension, density, and intensity; Second, adopt the Boosting supervised learning algorithm to automatically learn the discriminative differences between trunk point clouds and non-trunk point clouds, and obtain a high-precision trunk point cloud detector via feature fusion; Third, based on the recognition results of trunk point clouds, segment and label each individual street tree trunk. This proposed method can accurately extract street tree trunks and is applicable to the field of street tree resource survey.

搜集汇总

背景与挑战

背景概述

该数据集提供了一种基于移动激光扫描点云数据的行道树树干提取方法，包括构建点云标注数据集、提取多维度特征和应用监督学习算法进行树干识别。数据集属于人工林重大灾害防控关键技术研究项目，适用于行道树资源调查领域。