The Volume And Leaf Area Estimation Of The Forestry Tree Crown Based On Laser Surveying And Mapping

Estimating tree canopy volume and leaf properties is crucial for understanding the respiratory,transpiration,and photosynthetic processes of trees,as well as evaluating plant or forest biophysical parameters.Li DAR(Light Detection and Ranging)has been proven effective in various surveying projects,and with the development of relevant theoretical and technical advancements,the accuracy and efficiency of Li DAR surveying have continuously improved,better meeting the practical surveying needs and achieving favorable results.The work conducted in this article includes the following:(1)Based on the technical principles and characteristics of Li DAR surveying,the data of the selected study area were processed,including steps such as branch and leaf separation,segmentation,smoothing,data registration,and denoising.In the branch and leaf separation step,this article improved the Euclidean clustering segmentation method and introduced the concept of spatial indexing to handle the processing of branches and leaves in the fragmented tree point cloud.(2)Proper estimation of gap proportions is critical for quantifying canopy structure variables such as leaf area and apparent clumping index,which are used to correct for land-atmosphere interactions.However,the porosity measured by optical sensors is influenced by plant elements and scattered radiation from the ground surface.In this paper,a simple one-dimensional,reversible,two-way transport model is proposed to eliminate the effect of scattering in porosity measurements,which can simulate illumination well.Canopy volumes are usually calculated using conventional methods,but these are time consuming and have limited accuracy.In order to overcome these shortcomings,this paper adopts the 3D convex hull method and the method of calculating canopy volume from the elevation model(CHM),and canopy volume measurements with high accuracy are obtained from the 3D model.In addition,according to the statistical results,the time for the improved convex hull algorithm to calculate the projected area and volume of the tree crown is only 21%of that of the voxel simulation method,and 38%of that of the triangular network modeling method.(3)In order to improve the extraction methods of terrestrial Li DAR scanning in canopy characterization,this article constructed a methodological paradigm that combines physics and statistics to derive leaf profiles,leaf area index(LAI),and leaf angle distribution(LAD).By probabilistic modeling of the interaction between Li DAR and vegetation,a maximum likelihood estimation(MLE)method for vegetation parameters was proposed.Unlike traditional gap-based algorithms,MLE explicitly considers the geometric shape of Li DAR scanning,fully utilizes the original Li DAR ranging data,and obtains both leaf outlines and LAD simultaneously.Compared with traditional gap analysis,the leaf density distribution and LAD obtained by MLE are more accurate.In addition,MLE has a solid statistical foundation,and the error intervals generated better reflect the uncertainty of the estimated crown parameters,avoiding the issue of ignoring this uncertainty,which is crucial for the reliability of vegetation scanning using Li DAR.The R~2between the leaf inclination angle and leaf azimuth angle obtained by the experiment and the actual measured value are all more than 0.9.