【摘要】 以湖南省马尾松、栎类2个树种（组）为研究对象，利用地基激光雷达扫描样木，获取树高、胸径、地径等林木参数，基于激光雷达点云数据构建树高曲线模型和材积模型，并以解析木调查实测数据为对照，检验点云数据提取的林木参数精度和评估林业数表模型的适用性。结果表明：1）地基激光雷达提取的林木参数与同口径解析木实测数据无显著差异，马尾松、栎类2个树种（组）点云数据提取的胸径估计精度均达98%以上，树高估计精度在92%以上，材积估计精度在90%左右；2）基于点云数据所建的二元立木材积模型、一元立木材积模型和地径材积模型拟合精度高，其中二元立木材积模型的R~2均达0.99、预估精度（P）超过98%，马尾松、栎类2个树种（组）的树高曲线模型的P达96%，与实测数据所建模型的拟合精度相近。由此可见，地基激光雷达技术可实现非破坏性、快速精准的林木参数提取，构建基于点云数据的高精度林业数表模型，为限制或禁止采伐情况下的林业数据获取和数表模型研建提供了一种新的解决方案。更多还原

【Abstract】 Taking two tree species(groups) of Pinus massoniana and Quercus spp. in Hunan Province as research objects, this study utilized terrestrial laser scanning(TLS) to obtain tree parameters such as tree height,diameter at breast height(DBH), and ground diameter. Tree height curve models and volume models were constructed based on TLS point cloud data. The accuracy of the tree parameters extracted from the point cloud data was tested, and the applicability of the forestry table models was evaluated by comparing with measured data from dendrometric analysis. The results indicated that: 1) there was no significant difference between the tree parameters extracted from TLS data and those obtained from dendrometric analysis. The estimation accuracy of DBH extracted from point cloud data for both Pinus massoniana and Quercus spp. reached above98%, while the estimation accuracy of tree height exceeded above 92%, and that of volume was around 90%.2) The binary standing timber volume models, univariate standing timber volume models, and ground diameter volume models built using point cloud data exhibited high fitting accuracy. The determination coefficients of the binary standing timber volume reached 0.99, with an estimation accuracy exceeding 98%. The tree height curve models for both two tree species(groups) had an estimation accuracy of 96%, matching the fitting accuracy of models constructed from measured data. In conclusion, terrestrial laser scanning technology enabled nondestructive, rapid, and precise extraction of tree parameters, allowing for the construction of highly accurate forestry table models based on point cloud data. This approach provides a new solution for forestry data acquisition and table model development under conditions where logging is restricted or prohibited.更多还原