| 基于地基激光雷达胸径提取的单木位置精确测量 |
| |
| 引用本文： | 梁晓军,庞勇,陈博伟.基于地基激光雷达胸径提取的单木位置精确测量[J].林业科学研究,2020,33(4):67-74. |
| |
| 作者姓名： | 梁晓军 庞勇 陈博伟 |
| |
| 作者单位： | 中国林业科学研究院资源信息研究所,国家林业和草原局林业遥感与信息技术重点实验室 北京100091 |
| |
| 基金项目： | 国家重点研发计划;国家自然科学基金 |
| |
| 摘 要： | 目的]为将单木位置匹配至更精确的树干中心处,本研究发展了一种基于地基激光雷达提取胸径中心位置的空间校正方法。方法]对高密度地基激光雷达点云数据,使用霍夫变换的方法提取单木胸径及圆心点,以外业调查的胸径数据作为精度控制基础,再用空间点校正方法将外业测量的样地单木相对空间位置匹配至提取的单木胸径中心点处,最后通过数字冠层高度模型特征分析实现单木位置的地理位置匹配。结果]以黑龙江佳木斯孟家岗林场为研究区,对比分析3种不同株数密度的落叶松样地,提取胸径误差在1 cm之内,高、中、低株数密度样地单木位置在空间点校正时胸径误差2 cm误差范围内位置正确匹配率分别为91%、92.5%、98.6%。全部匹配的外业调查单木相对空间位置误差控制在1 m之内,且与机载激光雷达数字冠层模型影像地理位置匹配误差在0.5 m内。结论]基于地基激光雷达提取胸径匹配单木空间位置的方法,极大地提高了单木空间位置测量精度。此方法的发展,不仅为局部样地单木分割等提供精确位置信息,也为大范围遥感数据提供可靠地面基础位置验证数据,是可靠的单木位置测量和多源数据匹配方法。
|
| 关 键 词： | 地基激光雷达 霍夫变换 胸径拟合 影像林隙特征 单木位置匹配 |
Accurate Measurement of Individual Tree Position Based on DBH Extraction of Terrestrial Laser Scanning |
| |
| LIANG Xiao-jun,PANG Yong,CHEN Bo-wei.Accurate Measurement of Individual Tree Position Based on DBH Extraction of Terrestrial Laser Scanning[J].Forest Research,2020,33(4):67-74. |
| |
| Authors: | LIANG Xiao-jun PANG Yong CHEN Bo-wei |
| |
| Institution: | (Research Institute of Forest Resource Information Techniques,Chinese Academy of Forestry,Key Laboratory of Forestry Remote Sensing and Information System ofNational Forestry and Grassland Administration,Beijing 100091,China) |
| |
| Abstract: | Objective] To develop a spatial correction method based on the position of DBH center extracted from terrestrial laser scanning in order to match the position of individual trees to center of trunk precisely. Method]Hough transform method was used to extract the DBH and circular center of individual trees from point cloud data of high-density terrestrial laser scanning in the sample plots. Then, based on the DBH data obtained from field survey,the relative spatial positions of individual trees were matched to the extracted DBH center of individual trees by the spatial point correction methods. Finally, the location matching of individual trees was realized by CHM feature analysis. Result] The trial plots were located at Mengjiagang Forest Farm in Jiamusi, Heilongjiang Province. The relevant results of three Larix olgensis Henry forest plots with different stem densities were compared and analyzed. The errors of extracted DBH were within 1 cm. And when the DBH error was within 2 cm, the correct matching rate of individual trees position in high, middle and low stem density plots were 91%, 92.5% and 98.6% respectively. Besides,the relative spatial position errors of all matched field surveys were within 1 m and the errors of geographical position matching with the airborne LiDAR CHM products were within 0.5 m. Conclusion] The method of extracting DBH from terrestrial laser scanning to match spatial positions of individual trees can greatly improve the precision of individual trees spatial position. It can both provide precise location information for individual trees segementation in local sample plot, and give reliable location verification information for large-scale remote sensing data. Overall, it is approved as a reliable method for individual trees position measurement and multi source data matching. |
| |
| Keywords: | Terrestrial Laser Scanning Hough transform DBH extract image gap characteristics individual tree position matching |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
