基于地基激光雷达的落叶松人工林枝条因子提取和建模

Extraction of branch factors and model construction for Larix plantation using terrestrial laser scanning(TLS)

地基激光雷达(TLS)可以实现从森林中无破坏收集数据。本文基于地基激光雷达数据通过点云处理软件以人机交互的方式获取了26株落叶松样木的1266组枝条信息,包括着枝高度、弦长、枝长、着枝角度、基径和弓高。枝条可提取的最大相对着枝高度的平均值为0.83。在所提取的枝条因子中,提取精度依次为着枝高度>弦长>枝长>基径(基径大于20 mm的枝条)>弓高,将树冠分为4部分后分析发现,随着冠层高度的增加,枝条密度呈升高趋势,枝条提取率和提取精度呈下降趋势。此外,由于枝条基径提取精度较低,以弦长、着枝高度、胸径和树高为自变量构建基径预测模型。对不同基径的实测值、提取值与模型预测值对比分析发现,枝条基径的预测精度大于提取精度。对于造材来说,最有价值的部分是树木中下部,本方法能够较准确地提取树木胸径树高和相对着枝高度0.8以下的枝条属性信息,提供构建木材质量模型所需要的参数。

Forest parameters could be estimated without damage using terrestrial laser scanning(TLS).In this study,1266 sets of branch parameters,including branch height,branch chord length,branch length,branch angle,branch diameter and the branch arc height,of 26 Larix trees were extracted based on TLS using point cloud processing software in the way of human-computer interaction.The average of extracted maximum relative height of the branches was 0.83.The accuracy of the extracted branch attributes followed the order of branch height>branch chord length>branch length>branchdiameter(with the branch diameter>20 mm)>branch arc height.After dividing the crown into four parts,it proved that the density of the branches increased while the extraction ratio and precision of branches decreased with the increase of crown height.Due to the low accuracy of branch diameter extraction,we developed a base diameter prediction model using the branch chord length,the height of branch,diameter at breast height and tree height as independent variables.The measured,extracted and model predicted values of different base diameter were compared,and discovered that accuracy of model prediction was greater than that of extraction by TLS.For timber production,the most worthy wood was the middle and lower section of the tree.The method could accurately extract tree height,diameter at breast height(DBH)and the branch attributes with the relative branch height of less than 0.8,providing the parameters that are necessary to develop a wood quality model.