摘要
林业资源是保持生态环境、维持生态平衡的重要资源,对单木参数信息的提取是林业资源调查的重要内容。利用地基三维激光点云提取单木参数具有效率高、速度快、节省人力物力等优点。在常用的Hough拟合圆柱法提取树高参数和Hough拟合圆法提取胸径参数的基础上,提出格网化拟合圆柱法和最小二乘拟合圆法分别提取单木树高参数和胸径参数,并以现场直接量测的树高和胸径参数为参考值进行对比。实验结果显示:与直接量测结果相比较,Hough变化拟合圆柱法和格网化拟合圆柱法得到的树高参数平均偏差分别为0.05m、0.003m,标准差分别为0.142m、0.002m,相关系数分别为0.87、0.99;Hough变化拟合圆法和最小二乘拟合圆法得到的胸径参数平均偏差分别为0.003 6m、0.002 5m,标准差分别为0.004m、0.003m,相关系数分别为0.60、0.91。结果表明,提出的格网化拟合圆柱法提取树高参数和最小二乘拟合圆法提取胸径参数能得到精度较高的单木参数信息。
Forest resources is one of the most important resources for ecological environment construction and balanced ecosystem.The extraction of individual tree information is a key substance for forestry resources survey.Based on Hough fitting cylinder method for extraction of individual tree height and Hough fitting cycle method for diameter at breast height(DBH),the gridding fitting cylinder method as well as least square fitting cycle method are present in this study.Comparing with actually measured values,it shows that for individual tree heights estimated from Hough fitting cylinder method and gridding fitting cylinder method,the mean deviations are 0.05 mand 0.003 m,respectively;the standard deviations are 0.142 m and 0.002 m,respectively;the correlation coefficients are 0.87 and 0.99,respectively.As for individual tree DBH,the mean deviations are 0.003 6 mand 0.002 5 m,respectively;the standard deviations are 0.004 mand 0.003 m,respectively;the correlation coefficients are 0.60 and 0.91,respectively.It indicates that there have good results for extraction of individual tree characteristics using the gridding fitting cylinder method and fitting cycle method.
作者
郭沈凡
顾波
奚冠凡
陆晓勇
GUO Shen-fan;GU Bo;XI Guan-fan;LU Xiao-yong(Nanjing Institute of Surveying,Mapping&Geotechnical Investigation,Co.Ltd,Nanjing Jiangsu210000,China;College of Civil Engineer of Nanjing ForestryUniversity,Nanjing Jiangsu210037,China)
出处
《现代测绘》
2019年第2期22-25,共4页
Modern Surveying and Mapping
基金
江苏省测绘地理信息科研项目(JSCHKY201903,JSCHKY201806)
关键词
地基Lidar
点云
树高
胸径
ground-based lidar
point cloud
tree height
DBH
