Individual Tree Crown Segmentation Directly from UAV-Borne LiDAR Data Using the PointNet of Deep Learning
Individual Tree Crown Segmentation Directly from UAV-Borne LiDAR Data Using the PointNet of Deep Learning
复制标题
使用深度学习的 PointNet 直接根据无人机搭载的 LiDAR 数据进行单棵树冠分割
DOI：
10.3390/f12020131
复制
发表时间：
2021-02-01
期刊：
影响因子：
2.9
通讯作者：
Yun, Ting
中科院分区：
文献类型：
--
作者：
Chen, Xinxin;Jiang, Kang;Yun, Ting
从扫描点云中准确地进行单棵树冠（ITC）分割是森林生物量监测和森林生态管理中的一项基本任务。激光探测与测距（LiDAR）作为森林测量的主流工具，正在推动森林数据采集模式的发展。在本研究中，我们采用了一种新颖的深度学习框架，直接处理属于四种森林类型（即苗圃基地、寺院园林、混交林和落叶林）的森林点云，以实现ITC分割。我们方法的具体步骤如下：首先，采用体素化策略，将从不同森林类型收集的包含各种树种的点云细分为许多体素。这些包含点云的体素被作为PointNet深度学习框架的训练样本，以便在体素尺度上识别树冠。其次，基于初始分割结果，我们利用与高度相关的梯度信息来精确描绘每棵树冠的边界。同时，将检索到的单棵树的树冠宽度与实地测量结果进行比较，以验证我们方法的有效性。在这四种森林类型中，我们的结果显示苗圃基地的表现最佳（树冠检测率r = 0.90；树冠宽度估计的R²>0.94，均方根误差（RMSE）<0.2米）。对于寺院园林和混交林，尽管它们森林结构复杂、树枝交叉繁杂且建筑类型不同，也取得了良好的表现，寺院园林的r = 0.85，R²>0.88，RMSE < 0.6米，混交林的r = 0.80，R²>0.85，RMSE < 0.8米。对于林地中树冠落叶分布的第四种森林样地类型，我们取得了r = 0.82，R²>0.79，RMSE < 0.7米的效果。我们的方法提出了一个受深度学习技术和计算机图形学理论启发的稳健框架，解决了ITC分割问题，并在各种森林条件下获取了森林参数。
Accurate individual tree crown (ITC) segmentation from scanned point clouds is a fundamental task in forest biomass monitoring and forest ecology management. Light detection and ranging (LiDAR) as a mainstream tool for forest survey is advancing the pattern of forest data acquisition. In this study, we performed a novel deep learning framework directly processing the forest point clouds belonging to the four forest types (i.e., the nursery base, the monastery garden, the mixed forest, and the defoliated forest) to realize the ITC segmentation. The specific steps of our approach were as follows: first, a voxelization strategy was conducted to subdivide the collected point clouds with various tree species from various forest types into many voxels. These voxels containing point clouds were taken as training samples for the PointNet deep learning framework to identify the tree crowns at the voxel scale. Second, based on the initial segmentation results, we used the height-related gradient information to accurately depict the boundaries of each tree crown. Meanwhile, the retrieved tree crown breadths of individual trees were compared with field measurements to verify the effectiveness of our approach. Among the four forest types, our results revealed the best performance for the nursery base (tree crown detection rate r = 0.90; crown breadth estimation R-2 > 0.94 and root mean squared error (RMSE) < 0.2m). A sound performance was also achieved for the monastery garden and mixed forest, which had complex forest structures, complicated intersections of branches and different building types, with r = 0.85, R-2 > 0.88 and RMSE < 0.6 m for the monastery garden and r = 0.80, R-2 > 0.85 and RMSE < 0.8 m for the mixed forest. For the fourth forest plot type with the distribution of crown defoliation across the woodland, we achieved the performance with r = 0.82, R-2 > 0.79 and RMSE < 0.7 m. Our method presents a robust framework inspired by the deep learning technology and computer graphics theory that solves the ITC segmentation problem and retrieves forest parameters under various forest conditions.