在传统森林计测中,由于树木冠层的形态各异、结构复杂,往往难以精确获取结构因子.本文以地基激光雷达为工具,通过对单木扫描获取点云数据,基于球极平面投影和Lambert方位角等面积投影法计算冠层孔隙度,运用不规则面投影法和体元法提取树冠的体积和表面积,并对结果进行了对比分析.选取北京林业大学校园内6株立木为研究对象,结果表明,对于冠层孔隙度,球极平面投影的结果均小于Lambert方位角等面积投影,平均误差为0.03;对于体积和表面积,运用不规则面投影法和体元法得到树冠体积相对误差为5.32％ ～ 12.43％,平均相对误差为9.29％,提取的树冠表面积的相对误差为1.40％ ～5.21％,平均相对误差为3.33％.2种方法得到的结果差别不是很显著.因此利用地基激光雷达获取点云数据,通过对单木一次扫描提取冠层孔隙度、树冠体积和表面积,为计算树木的三维绿量、生物量、光合作用能力等提供了更可靠的数据支持.%In traditional forest resources survey,it was hard to get tree structure parameters due to its different shapes and complex structures.Totally six stand trees were selected as a sample data in the campus of Beijing Forestry University.By using ground-based laser scanner as a tool to acquire single tree scanning point cloud data,the canopy gap fraction was calculated based on Lambert azimuthal equal area projection and stereographic projection;tree crown volume and surface area were extracted by two methods:irregular surface projection and voxel method by using cloud point data.The results showed that the average relative error of two methods for gap fraction was 0.03,and the extraction result using stereographic projection was smaller than that using Lambert azimuthal equal area projection;the relative error of crown volume of two kinds of extraction methods was 5.32％ ～ 12.43％,the average relative error was 9.29％,the relative error of the extracted crown surface area was 1.40％ ～ 5.21％,the average relative error was 3.33％.The difference between the results of the two methods were not very significant,therefore,the extraction of gap fraction,crown volume and surface area based on the use of the point cloud data can provide a more reliable data support for 3D green quantity,calculation of trees biomass,photosynthetic ability etc.
展开▼
