基于机载LiDAR全波形数据白桦林林分LAI反演研究  

A Inversion on Birch Forest LAI Based on Airborne LiDAR Full Waveform Data

作　　者：邢艳秋[1] 姚松涛[1] 尤号田[1] 田昕[2] 彭涛[3] 李梦颖[1] 谢杰[1] 闫灿[1] 

机构地区：[1]东北林业大学森林作业与环境研究中心,黑龙江哈尔滨150040 [2]中国林业科学研究院资源信息研究所,北京100091 [3]东北林业大学信息与计算机工程学院,黑龙江哈尔滨150040

出　　处：《西北林学院学报》2018年第1期11-18,共8页Journal of Northwest Forestry University

基　　金：国家重点基础研究发展计划项目(2013CB733404);林业公益性行业科研专项(201504319).

摘　　要：为提高森林叶面积指数(LAI)的估测精度,本研究以白桦林为研究对象,以机载激光雷达(LiDAR)全波形数据为研究数据,首先提出了机载LiDAR全波形数据读取与波形特征信息提取的相关算法,结合具体算法的实现分析出每条全波形对应的各波形分量的能量信息,然后依据波形能量信息在传统激光穿透指数(LPI)计算的基础上结合全波形数据的特点,计算出全波形激光穿透指数(LPIfi),最后获得样地体元激光穿透指数(LPIf-mean),用于估测森林林分LAI,并将估测结果与机载LiDAR离散回波点云数据估测森林LAI的结果进行了对比。结果表明,全波形样地体元激光穿透指数LPIf-mean与森林林分LAI之间的建模精度R^2=0.815,RMSE=0.105,预测精度R^2=0.864,RMSE=0.139,同时在同等样地尺度下,全波形数据返回脉冲能量信息估测森林LAI的精度要高于离散回波点云数据的,因而,基于机载LiDAR全波形数据能够实现森林林分LAI的高精度反演。可以为进一步森林生态参数模拟与估测提供高精度的基础数据,弥补和提供了机载LiDAR全波形数据估测森林LAI的方法和思路。In order to improve the estimation accuracy of forest leaf area index （LAI）, this paper took birch forest as the research object and the full-waveform data of airborne light detection and ranging （LiDAR） as the research data. Firstly,the algorithm of airborne LiDAR full-waveform data reading and waveform fea- ture information extraction were proposed to analyze the energy information of each waveform component corresponding to each full-waveform,and then based on the waveform energy information in the traditional laser penetrating index （LPI） calculation,the whole waveform data （LPIf ） were obtained, which were used to estimate the LAI of forest stands and to estimate the results with the airborne LiDAR discretiza- tion （LPIfi）. The echo point cloud data were used to estimate the results of forest LAI. The results showed that the modeling accuracy R2 = 0. 815, the RMSE= O. 105, the prediction accuracy was R2 ： 0. 864,RMSE =0. 139, and the same accuracy was obtained at the same size as the plot voxel laser penetration index LPIf and the forest stand LAI. The full-waveform data return pulse energy information to estimate the forest LAI accuracy was higher than the discrete echo point cloud data. Therefore, based on the airborne LiDAR,full-waveform data could realize the high precision inversion of the forest stand LAI,which could provide high precision basic data for the further simulation and estimation of forest ecological parameters, to make up and provide the airborne LiDAR full-waveform data to estimate the forest LAI methods.

关 键 词：机载激光雷达 全波形数据 样地体元激光穿透指数 白桦林 叶面积指数 

分 类 号：S792.153[农业科学—林木遗传育种]