植物生态学报 ›› 2015, Vol. 39 ›› Issue (7): 694-703.DOI: 10.17521/cjpe.2015.0066

• 研究论文 • 上一篇    下一篇

集成高分辨率UAV影像与激光雷达点云的亚热带森林林分特征反演

许子乾1,2, 曹林1,2, 阮宏华1,2,*(), 李卫正3, 蒋圣4   

  1. 1南京林业大学生物与环境学院, 南京 210037
    2南方现代林业协同创新中心, 南京 210037
    3南京林业大学现代分析测试中心, 南京 210037
    4南京师范大学地理科学学院, 南京 210046
  • 出版日期:2015-07-01 发布日期:2015-07-22
  • 通讯作者: 阮宏华
  • 作者简介:

    *作者简介:E-mail:dengchuanyuan@163.com

  • 基金资助:
    国家重点基础研究发展计划项目(2012CB416904)、国家自然科学基金(41401440)和江苏生物学优势学科建设项目

Inversion of subtropical forest stand characteristics by integrating very high resolution imagery acquired from UAV and LiDAR point-cloud

XU Zi-Qian1,2, CAO Lin1,2, RUAN Hong-Hua1,2,*(), LI Wei-Zheng3, JIANG Sheng4   

  1. 1College of Biology and Environment, Nanjing Forestry University, Nanjing 210037, China
    2Co-innovation Center for Sustainable Forestry in Southern China, Nanjing 210037, China
    3Advanced Analysis and Testing Center of Nanjing Forestry University, Nanjing 210037, China
    4College of Geography Science, Nanjing Normal University, Nanjing 210046, China
  • Online:2015-07-01 Published:2015-07-22
  • Contact: Hong-Hua RUAN
  • About author:

    # Co-first authors

摘要:

该研究集成高分辨率无人机(UAV)影像和激光雷达(LiDAR)点云数据估算亚热带天然次生林林分基本特征变量。首先, 基于LiDAR点云和反距离加权插值法构建林下高精度数字高程模型(DEM); 然后利用UAV影像对序列构建植被冠层上层三维点云, 并借助DEM进行高度信息归一化, 提取高度和冠层点云密度相关的特征变量; 最后, 构建预测模型并估算Lorey’s高、林分密度、胸高断面积、蓄积量。结果表明: 联合提取的特征变量与Lorey’s高的敏感性最高, 蓄积量次之, 林分密度和胸高断面积最低; 利用UAV灵活快速的手段获取森林冠层信息, 辅以高精度LiDAR数据获取的地形信息, 两者互补实现一种可重复的快速、廉价和灵活的林分特征的反演方式。

关键词: 无人机, 激光雷达, 点云, 林分特征

Abstract: Aims We applied the integrated very high resolution imagery acquired from Unmanned Aerial Vehicles (UAV) and Light Detection and Ranging (LiDAR) point-loud data to estimate the stand characteristics of a naturally- regenerated forest in a subtropical area. Methods The high precision digital elevation model (DEM) of the forest was constructed base on LiDAR point-cloud and the inverse distance weighted interpolation method. The 3D point-cloud of forest canopy layer was constructed from UAV image pairs, with information from DEM height information normalization, for canopy height and density. With the above effort, we developed a prediction model to estimate Lorey’s height, stand density, basal area, and volume. Important findings The quantitative metrics generated from this study appeared very sensitive to Lorey’s height, followed by volume and basal area. Using UAV as a flexible and rapid method for generating forest canopy characteristics, combined with topographic information from high precision LiDAR data, seems a viable, rapid, inexpensive, and flexible method in canopy research.

Key words: LiDAR, point cloud, stand characteristics, UAV