| 面向人工林智慧监测的近地面激光雷达单木分割重采样校正 |
| Research on Individual Tree Segmentation Resampling Correction Based on Terrestrial LiDAR for Artificial Forest Intelligent Monitoring |
| 投稿时间：2025-10-08 修订日期：2025-10-29 |
| DOI：10.19316/j.issn.1002-6002.2025.S1.12 |
| 中文关键词: 单木分割 激光雷达 林业监测 空间重采样 后处理校正 |
| 英文关键词:individual tree segmentation laser radar forestry monitoring spatial resampling post-processing correction |
| 基金项目: |
|
| 通讯作者:郭伟* 福建省环境监测中心站, 福建 福州 350003 |
| 摘要点击次数: 347 |
| 全文下载次数: 138 |
| 中文摘要: |
| 针对机载激光雷达(ALS)数据在人工林单木分割中存在的漏检与欠分问题,提出一种基于空间规则约束的重采样校正方法。以某桉树人工林为试验区,在传统单木分割算法得出的初始结果的基础上,结合椭圆环形区域插值与动态迭代,生成更加符合真实空间分布的单木位置点集。实验结果表明:所提方法对单木分割结果产生了正向校正作用。在训练与验证样地中,分割算法的召回率与F1分数均获得大幅提升,同时显著改善了林分密度和空间分布重建精度,充分证明了该方法的有效性和实用性。该方法解决了ALS数据中小树漏分割、点群偏移等问题,为人工林智慧监测提供了创新性技术方案。 |
| 英文摘要: |
| Aiming at the problem of missing detection and under-segmentation of Airborne Laser Scanning (ALS) data in individual tree segmentation of artificial forest,this paper proposes a resampling correction method based on spatial rule constraints.Taking a eucalyptus plantation as the experimental area,based on the initial results obtained by the traditional individual tree segmentation algorithm,combined with the ellipse annular region interpolation and dynamic iteration,a individual tree position point set that is more in line with the real spatial distribution is generated.The experimental results show that the proposed method has a positive correction effect on the individual tree segmentation results.In both training and verification plots,the recall rates and F1-scores of the segmentation algorithm were greatly improved,while the stand density and spatial distribution reconstruction accuracy were significantly improved,which fully proved the effectiveness and practicability of the method.This method solves the problems of small tree missing segmentation and point group offset in ALS data,providing an innovative technical solution for artificial forest intelligent monitoring. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
