欢迎访问林业资源管理

林业资源管理 ›› 2007, Vol. 0 ›› Issue (6): 73-77.

• 科学技术 • 上一篇    下一篇

LIDAR技术在林业调查中的应用研究

张彦林1,2, 冯仲科1, 杨伯钢1, 马友平1, 姚山1   

  1. 1.北京林业大学测绘与3S技术中心,北京100083;
    2.甘肃林业职业技术学院,甘肃 天水 741020;
    3.北京市测绘设计研究院,北京100038
  • 收稿日期:2007-05-16 修回日期:2007-09-17 发布日期:2020-12-10
  • 作者简介:张彦林(1963-)男,甘肃天水人,副教授,博士生,主要研究方向为3S技术在林业中应用。Email:cato2005@sina.com;责任作者:冯仲科,教授,博士生导师,主要研究方向为3S技术。Email:fengzhongke@126.com

Applications of LIDAR Technology in Forestry Investigation

ZHANG Yanlin1,2, FENG Zhongke1, YANG Bogang3, MA Youping1, YAO Shan1   

  1. 1. Institute of GIS, RS & GPS, Beijing Forestry University, 100083, P. R. China;
    2. Gansu Forestry Technological College, Tianshui,741020, Gansu,P.R. China;
    3. Beijing Research Institute of Surveying and Designing, Beijing 100038,China
  • Received:2007-05-16 Revised:2007-09-17 Published:2020-12-10

摘要: 以LIDAR测量树高为研究内容,利用全站仪对研究区选定的37株样木的树高和材积进行精准测量,获取训练样本实测数据。通过BP神经网络分析,建立了LIDAR测量树高与实测树高、材积的拟合函数方程,实现了用LIDAR树高数据计算单木材积的过程。从拟合的结果来看,树高相关系数R为0.993 2,平均残差平方和为0.086 312。材积相关系数R为0.974 7,平均残差平方和为0.000 310 5,达到了林业调查精度的要求,在林业数表的编制方面是一个比较好的范例。

关键词: LIDAR, 树高, 材积, BP神经网络, 拟合函数方程

Abstract: This paper studied measurement of tree heights by using LIDARtechnology. The total station precisely measured height and volume of the 37 selected sample trees, which served as measured sample data. With the analysis ofBPNeural Networks, function equation was established to fit LIDAR images to real-measured tree height and volume data, which processed realization of using LIDARdata to calculate individual tree volume. It can be drawn from the fitting result that correlative coefficient of height was 0.993 2 and mean residual sum of squares was 0.086 312, correlative coefficient of volume was 0.974 7 and mean residual sum of squares was 0.000 310 5, this result metthe requirement of forestry investigation precision. It also set a good examplefor forestry tables making.

Key words: LIDAR, tree heights, tree volumes, BP Neural Networks, fitting function equation

中图分类号: