基于UAV激光雷达和多光谱数据的森林单木生物量监测技术研究

Accurate estimation of forest biomass at individual tree level is the foundation of timber industry and forest management, and play an important role in explaining ecological issues and process at individual tree level. More importantly, remote sensing applications of biomass estimation in the individual tree level bases are significant for the development of theory and technology in the field of ecological remote sensing. The potential of UAV LiDAR to monitor forest biomass has been proved in the literature. However, it is still a challenge to estimate forest biomass at the individual tree level based on UAV LiDAR data, which is mainly reflected in low accuracy of single tree detection and segmentation for broadleaf forest and the difficulty of DBH estimation in tree level. Therefore, this project proposed to create a new method to detect and segment single tree for broadleaf forest (using poplar plantation as an example); to explore algorithm an accuracy for monitoring DBH, tree height and crown width at individual tree level for broadleaf forest based on the combination of UAV LiDAR and multispectral data; and future on the foundation of systematically analyzing the impact of forest age, stand density and herbs on the accuracy for monitoring single tree biomass, to establish a new methodology to monitor broadleaf forest biomass at individual tree level via the combination of UAV LiDAR and multispectral data. This proposed research have the potential to improve and enrich the theory and technology of ecological remote sensing in the terms of research contents and methods, which has significant theoretical value; meanwhile, the proposed research results would provide the fundamental information and technique support for forest management decision maker, which has important application value.

在单木尺度上精确估测森林生物量是工业用材及森林管理的基础，也是解释个体层面上生态问题和生态过程的必要环节和重要研究内容。开展单木尺度生物量的遥感研究更是对生态遥感理论和技术的发展有重要的意义。无人机激光遥感监测森林生物量的能力已得到验证，但其在单木尺度上的研究仍存在难点与挑战，主要表现在：对阔叶林单木识别与分割精度较低、单木尺度树木胸径的难以精确估测。因此，本项目拟以杨树人工林为研究对象，创建阔叶林区单木识别与单木分割的新方法；探索无人机激光点云数据和无人机多光谱影像在单木尺度上监测阔叶林胸径、树高和冠幅的算法及精度；在系统分析林龄、林分密度和林下草本对监测精度影响的基础上，最终建立结合无人机激光和多光谱数据的监测森林单木生物量的新途径。本项研究可从研究方法和研究内容上丰富和完善生态遥感理论，有重要的理论价值；同时研究结果也可为森林资源的科学管理提供方法支撑和决策依据，具有显著的应用价值。

在单木尺度上精确估测森林生物量是工业用材及森林管理的基础，开展单木尺度生物量的遥感研究更是对生态遥感理论和技术的发展有重要的意义。本项目拟以杨树人工林为研究对象，创建阔叶林区单木识别与单木分割的新方法；探索无人机激光点云数据和无人机多光谱影像在单木尺度上监测阔叶林胸径、树高和冠幅的算法及精度；在系统分析林龄、林分密度和林下草本对监测精度影响的基础上，最终建立结合无人机激光和多光谱数据的测算阔叶林单木参数的新途径。本项研究可从研究方法和研究内容上丰富和完善生态遥感理论，有重要的理论价值；同时研究结果也可为森林资源的科学管理提供方法支撑和决策依据，具有显著的应用价值。