Aiming to develop a new tree biomass estimation model that is adaptable to airborne observations of forest canopies by unmanned aerial vehicles (UAVs), we applied two theories of plant form; the pipe model theory (PMT) and the statical model of plant form as an extension of the PMT for tall trees. Based on these theories, tree biomass was formulated using an individual tree canopy height model derived from a UAV. The advantage of this model is that it does not depend on diameter at breast height which is difficult to observe using remote-sensing techniques. We also proposed a treetop detection method based on the fractal geometry of the crown and stand. Comparing surveys in plantations of Japanese cedar (Cryptomeria japonica D. Don) and Japanese cypress (Chamaecyparis obtusa Endl.) in Japan, the root mean square error (RMSE) of the estimated stem volume was 0.26 m³ and was smaller than or comparative to that of models using different methodologies. The significance of this model is that it contains only one empirical parameter to be adjusted which was found to be rather stable among different species and sites, suggesting the wide adaptability of the model. Finally, we demonstrated the potential applicability of the model to light detection and ranging (LiDAR) data which can provide vertical leaf density distribution.

中文翻译：

---

应用管道模型理论并适用于无人机衍生冠层高度模型的树生物量估计模型

为了开发一种新的树木生物量估算模型，该模型适用于无人飞行器（UAV）对森林冠层的空中观测，我们应用了两种植物形态理论：管道模型理论（PMT）和植物形态的静态模型是高大树木PMT的扩展。基于这些理论，使用源自无人机的单个树冠高度模型来制定树木生物量。该模型的优点是它不依赖于乳房高度处的直径，而使用遥感技术很难观察到直径。我们还提出了一种基于树冠和支架的分形几何形状的树梢检测方法。比较日本雪松（Cryptomeria japonica D. Don）和日本柏（Chamaecyparis obtusa）人工林的调查Endl。）在日本，估计茎干的均方根误差（RMSE）为0.26 m ³，小于或等于使用不同方法的模型的均方根误差。该模型的意义在于，它仅包含一个要调整的经验参数，发现该参数在不同物种和地点之间相当稳定，这表明该模型具有广泛的适应性。最后，我们证明了该模型对光检测和测距（LiDAR）数据的潜在适用性，该数据可以提供垂直叶片密度分布。