基于无人机影像估算林分树高的研究

Estimation of Tree Height in Forest Stands based on UAV Imagery

为了迅速获取森林资源调查的树高因子,以南宁树木园坛里管理区为研究对象,基于无人机影像数据,结合实地调查,采用PostFlight Terra 3D软件处理得到研究区的DOM、DEM和DSM,利用ArcGIS空间分析技术估测树高;比较分析实测样木树高与估测树高精度,建立实测样木树高与估测树高相关模型并估测林分平均高。结果表明:通过无人机影像提取的树高与实测树高比较,总体平均精度达90.9%;提取树高与实测树高存在较强相关性,模型树高的平均精度达93.2%;估测林分平均高准确率达97.2%。该树高参数自动化提取方法,能够达到相关实践要求,可以在一定程度上替代人工实测,在人工林中具有广阔的应用前景。

In order to quickly obtain the tree height factor for forest inventory,the Tanli management area of Nanning Arboretum was used as the research object.Based on the UAV image data,combined with the field survey,the DOM,DEM and DSM of the study area were obtained by processing with PostFlight Terra 3D software,and the tree height was estimated by using ArcGIS spatial analysis technology.The measured sample tree height was compared with the estimated tree height for accuracy.The results showed that the tree height of the sample trees was estimated by using UAV remote images,with overall average accuracy reaching 90.9%.A close correlation between the extracted tree height and the measured tree height existed,and the average accuracy of the model tree height is 93.2%,with the accuracy of average height of stand reaches 97.2%.The tree height parameters is automatically extracted,meeting the relevant practical requirements and can replace manual measurement to a certain extent.