[1]熊妮娜,王佳.基于地基激光雷达的活立木材积提取算法[J].林业工程学报,2020,5(06):143-148.[doi:10.13360/j.issn.2096-1359.202001035]
 XIONG Nina,WANG Jia.Extraction algorithm for stand volume using ground-based laser scanner[J].Journal of Forestry Engineering,2020,5(06):143-148.[doi:10.13360/j.issn.2096-1359.202001035]
点击复制

基于地基激光雷达的活立木材积提取算法()
分享到:

《林业工程学报》[ISSN:2096-1359/CN:32-1862/S]

卷:
5
期数:
2020年06期
页码:
143-148
栏目:
森林工程与土建交通
出版日期:
2020-11-01

文章信息/Info

Title:
Extraction algorithm for stand volume using ground-based laser scanner
文章编号:
2096-1359(2020)06-0143-06
作者:
熊妮娜1王佳23
1.北京市城市管理研究院,北京100028;2.北京林业大学精准林业北京市重点实验室,北京100083;3.北京林业大学测绘与3S技术中心,北京100083
Author(s):
XIONG Nina1 WANG Jia23
1. Beijing City Management Research Institute, Beijing 100028, China; 2. Beijing Key Laboratory of Precision Forestry, Beijing Forestry University, Beijing 100083, China; 3. Institute of GIS, RS & PS, Beijing Forestry University, Beijing 100083, China
关键词:
地基激光雷达点云数据立木材积改进Hough变换
Keywords:
groundbased laser scanner point cloud data stand volume improved Hough transform method
分类号:
S758
DOI:
10.13360/j.issn.2096-1359.202001035
文献标志码:
A
摘要:
传统的立木材积主要依靠胸径尺、测高器等工具对标准样地进行调查,效率慢、精度低。提出一种利用地基激光雷达立木材积提取算法:首先对样木进行扫描,获取点云数据,在数据处理软件中手工提取单木的点云数据,进而应用改进Hough变换方法,对树干某高度处点云数据作切片处理,投影后转为栅格图像,利用Hough变换方法提取拟合圆,然后将该拟合圆的直径视为该处树干直径,通过提取不同高度处树干直径,以断面积区分求积法计算树干材积。本研究以56棵山杨树为例,利用地基激光雷达提取材积与全站仪提取材积做对比,结果表明:两种方法平均相对误差为5.86%,说明两种方法精度十分接近,两种方法之间相关系数R为0.998,决定系数为0.996,所建立的一元线性模型的回归系数标准误差为0.022,模型回归系数较好,说明两种方法具有显著的相关性。因此利用地基激光雷达提取材积可以作为立木材积测算方法。
Abstract:
Single tree volume is the basic unit of the stand volume, which is the main basis for achieving the volume measurement result. The traditional single tree volume mainly relies on tools such as breast diameter ruler and altimeter to survey standard plots, which uses the diameter at breast height (DBH) and height of standard timber to query the volume table for estimation. This method is inefficient due to the effect of many humans interference factors, which greatly affects the accuracy of the volume table. This research proposes an algorithm for extracting timber products using groundbased laser scanner. Firstly, the samples are scanned to obtain point cloud data, and the point cloud data of the single tree is manually extracted in the data processing software. Then, an improved Hough transform method is applied to the tree trunk. The point cloud data at a certain height is sliced. After the projection, it is converted into a raster image. The fitted circle is extracted using the Hough transform method, and then the diameter of the fitted circle is regarded as the trunk diameter at that location. By extracting the trunk diameter at different heights, the trunk volume is calculated using the crosssectional area division quadrature method. In this study, 56 poplar trees are used as examples, and with the DBH measured directly on site as reference values, the DBHs extracted by the groundbased laser scanner are analyzed. The average relative error of the DBH has high accuracy of 1.81%. The volume of groundbased lidar is used to compare the volume of the total station with that of the total station. The results show that the average relative error of the two methods is 5.86%, indicating that the accuracy of the two methods is very close. The correlation coefficient R is 0.998, the coefficient of determination is 0.996, and the standard error of the regression coefficient of the established univariate linear model is 0.022. The regression coefficient of the model is good, indicating that the two methods have significant correlation and that groundbased laser scanner can be a single tree volume calculation method. The proposed measurement method has a wide application prospect in the field of forest resources monitoring.

备注/Memo

备注/Memo:
收稿日期:2020-01-22? ? 修回日期:2020-05-12
基金项目:北京市自然科学基金(8182038);国家自然科学基金(31870713);中央高校科研基本业务经费项目(2015ZCQ-LX-01)。
作者简介:熊妮娜,女,高级工程师,研究方向为测绘工程技术在林业、生态环境中的应用。Email:631821336@qq.com
更新日期/Last Update: 2020-11-11