欢迎访问林草资源研究

林业资源管理 ›› 2021›› Issue (2): 99-109.doi: 10.13466/j.cnki.lyzygl.2021.02.014

• 技术应用 • 上一篇    下一篇

基于背包式激光雷达的天山云杉林单木因子估测

邱琴1(), 张文革2, 王蕾3, 曹姗姗4, 孙伟1,4()   

  1. 1.新疆农业大学 计算机与信息工程学院,乌鲁木齐 830052
    2.新疆天山西部国有林管理局,新疆 伊宁 835000
    3.新疆林业科学院现代林业研究所,乌鲁木齐 830052
    4.中国农业科学院农业信息研究所,北京 100081
  • 收稿日期:2021-02-09 修回日期:2021-03-25 出版日期:2021-04-28 发布日期:2021-06-03
  • 通讯作者: 孙伟
  • 作者简介:邱琴(1996-),女,四川达州人,在读硕士,研究方向:农林信息智能分析。Email: 1207788620@qq.com
  • 基金资助:
    新疆维吾尔自治区自然科学基金面上项目(2018D01A20)

Estimation of Single Wood Factor of Picea schrenkiana var.tianshanica Forest Based on Backpack LiDAR

QIU Qin1(), ZHANG Wenge2, WANG Lei3, CAO Shanshan4, SUN Wei1,4()   

  1. 1. Computer and Information Engineering College,Xinjiang Agricultural University,Urumqi,Xinjiang 830052,China
    2. State owned Forest Administration Bureau of Western Tianshan,Yining,Xinjiang 835000,China
    3. Institute of Mordern Forestry,Xinjiang Academy of Forestry Science,Urumqi 830052,China
    4. Agricultural Information Institute,the Chinese Academy of Agricultural Sciences,Beijing 100081,China
  • Received:2021-02-09 Revised:2021-03-25 Online:2021-04-28 Published:2021-06-03
  • Contact: SUN Wei

摘要:

单木因子高精度无损快速估测对森林资源监测和评估至关重要,背包式激光雷达在获取森林三维结构参数方面具有良好的应用潜力。以天山云杉林为研究对象,利用背包式激光雷达扫描样地获取点云数据进行单木分割识别和单木胸径、树高及冠幅面积等因子估测,以地面实测结合目视解译数据作为参照,进行精度评价及相关性检验。结果表明:利用背包式激光雷达数据进行单木分割的单木分割精度F值均大于0.9,精确率和召回率均值分别为0.96和0.90,识别率平均值为86.61%;单木胸径和树高估测结果决定系数R2均大于0.90,胸径均方根误差RMSE均值为1.11,树高的为1.05;单木冠幅面积估测结果决定系数R2均大于0.80,均方根误差RMSE均值为3.21。可见,使用背包式激光雷达能够实现对单木胸径、树高参数的高精度提取。

关键词: 背包雷达, 天山云杉, 单木参数, 胸径, 树冠

Abstract:

High precision lossless rapid estimation of single wood factor is very important for forest resource monitoring and evaluation.Backpack lidar has good application potential in obtaining 3D forest structure parameters.The study takes Picea schrenkiana var.tianshanica forest as the research object,trying to obtain the point cloud data by backpack LiDAR scanning sample for single wood segmentation and estimation of single wood DBH,tree height and crown area.The accuracy evaluation and correlation test are carried out with the reference of ground measurement and visual interpretation data.The results show that:the precision F values of single wood segmentation using backpack LiDAR data are greater than 0.90,the average accuracy and recall are 0.96 and 0.90,respectively.The average recognition rate is 86.61%.R2 coefficient of determination of DBH and tree height is greater than 0.90,the average value of the DBH root mean square error is 1.11 and the height of the tree is 1.05;R2 coefficient of determination for area estimation of single crown is greater than 0.80,the average value of the root mean square error is 3.21.Therefore,using backpack LiDAR can achieve high precision extraction of single wood DBH and tree height parameters.

Key words: backpack radar, Picea schrenkiana var.tianschanica, single wood parameter, DBH, crown

中图分类号: