Supplementary Data and Code: Forest variables from LiDAR: Drone flight parameters impact the detection of tree stems and diameter estimates

The dataset consists of five individual LiDAR point cloud of a part of the LMU managed forest, located close to Landshut Germany. The data was recorded in May and July of 2024 and was used to analyse how different flight parameters influence the data for the estimation of forest structural parameters. The LiDAR data was collected using a GeoSun GS-100C+ LiDAR Scanner (Wuhan Geosun Navigation Technology Co., Ltd) on a DJI Matrice M300 Drone (SZ DJI Technology Co., Ltd.) and processed using Geosun's software shipped with the device. Further, all datasets were corrected for intensity and geolocation. Naming: H indicates flight height in meters, S flight speed in meters per second divided by 10, and O the overlap in percent, N denotes the flight being performed at night. The data was used to develop new methods for the delineation of stems from the point cloud, the estimation of understory vegetation, and the estimation of tree DBH values. We have found the best functioning of both stem delineation and DBH estimation under the densely overlapped data of H50 S60 O60. The added code for R presents the three methods for stem delineation, DBH estimation, and understory vegetation segmentation.该数据集包含五组独立的激光雷达（LiDAR）点云数据，采集自德国兰茨胡特（Landshut）附近由LMU管理的一片森林区域。数据采集于2024年5月与7月，旨在探究不同飞行参数对森林结构参数反演所用激光雷达数据的影响规律。本次激光雷达数据由搭载于大疆经纬M300无人机（DJI Matrice M300，深圳大疆创新科技有限公司）的GeoSun GS-100C+激光雷达扫描仪（武汉Geosun导航技术有限公司）采集，并使用设备配套的Geosun软件完成数据处理。此外，所有数据集均已完成强度校正与地理坐标校正。 数据命名规则如下：字母H代表飞行高度（单位：米），S代表飞行速度（单位：米/秒，实际值为标注值除以10），O代表重叠率（单位：%），字母N则表示该次飞行在夜间进行。 本数据集被用于研发从点云中提取树干轮廓、估算林下植被参数以及测算树木胸径（DBH，Diameter at Breast Height）的新方法。研究发现，在飞行参数设置为H50 S60 O60的高密度重叠点云数据场景下，树干轮廓提取与胸径估算的综合表现最佳。随数据集附带的R语言代码实现了上述三种方法，分别为树干轮廓提取、胸径估算与林下植被分割。