基于改进凸包算法的树冠轮廓点提取与体积计算
|
摘要
针对城市中树木数量多、树冠结构复杂、形态多变、难以测量等特点,为解决传统凸包算法在提取树冠轮廓时误差较大的问题,并实现树冠轮廓点提取和树冠体积自动计算,提出迭代渐进的凸包算法。以三维激光扫描仪获取的点云为原始数据,基于所提出的算法获取树冠轮廓点,结合格林公式及不规则台体体积法计算树冠体积。为验证算法的准确性,采用人工交互的方式提取树冠外轮廓点计算树冠体积,以此为参照对迭代渐进的凸包算法进行相关性与均方根误差的验证,并将该算法分别与几何体模型法、体元模拟法、Graham扫描线法计算的树冠体积进行对比分析,结果表明,迭代渐进的凸包算法计算出的树冠体积更接近于树冠的真实体积,为树冠信息的精细提取与建模提供一种新的方法。
For the number of trees in the city,the tree canopy structure is complex,shape-changing,difficult to measure and so on,In order to solve the problem that the traditional convex hull algorithm has a large error in extracting the crown and to complete the crown contour point extraction and crown volume automatic calculation,the iterative incremental convex hull algorithm is put forward,The data obtained with the 3 Dlaser scanner is used as the original data to get crown contour points based on the proposed algorithm.And the crown volume is calculated by combining the Green formula and the irregular bulk volume method.In order to verify the accuracy of the algorithm,the crown contour is extracted by artificial interaction.Based on this,the correlation and root mean square error of the iterative convex hull algorithm are verified.The algorithm is compared with the volume of the canopy calculated by geometric model method,voxel simulation method and Graham scanning line method respectively.The result shows that the crown volume by the iterative incremental convex hull algorithm calculates closer to the true volume of the canopy,which can provide a new method for extracting a fine crown Information and building the tree model.
For the number of trees in the city,the tree canopy structure is complex,shape-changing,difficult to measure and so on,In order to solve the problem that the traditional convex hull algorithm has a large error in extracting the crown and to complete the crown contour point extraction and crown volume automatic calculation,the iterative incremental convex hull algorithm is put forward,The data obtained with the 3 Dlaser scanner is used as the original data to get crown contour points based on the proposed algorithm.And the crown volume is calculated by combining the Green formula and the irregular bulk volume method.In order to verify the accuracy of the algorithm,the crown contour is extracted by artificial interaction.Based on this,the correlation and root mean square error of the iterative convex hull algorithm are verified.The algorithm is compared with the volume of the canopy calculated by geometric model method,voxel simulation method and Graham scanning line method respectively.The result shows that the crown volume by the iterative incremental convex hull algorithm calculates closer to the true volume of the canopy,which can provide a new method for extracting a fine crown Information and building the tree model.
引文
[1]林树森.城市增长与城市发展[J].城市规划,2011(11):11-18.
[2]林树森.城市道路交通长期规划应突出人居环境理念[J].城市与区域规划研究,2015(3):120-131.
[3]李杨,李秀峰.基于点云数据的树木骨架线提取研究[J].科技创新与生产力,2017(6):53-55.
[4]管西鹏.树木三维点云数据分析与建模技术研究[D].长沙:中南林业科技大学,2015.
[5]李宏星,欧阳玉华.基于三维激光扫描点云的树冠面积快速精准计算方法[J].绿色科技,2015(6):72-74.
[6]梁子瑜.基于TLS点云数据的林分调查因子测定及收获估计[D].南京:南京林业大学,2015.
[7]王祺,胡洪,吴艳兰,等.基于点云数据的树冠体积自动求算方法[J].西北林学院学报,2017,32(2):242-246.
[8]臧克.基于Riegl三维激光扫描仪扫描数据的初步研究[J].首都师范大学学报(自然科学版),2007,28(1):77-82.
[9]熊妮娜,王佳,罗旭,等.一种基于三维激光扫描系统测量树冠体积方法的研究——以油松为例[J].北京林业大学学报,2007(增2):61-65.
[10]韦雪花,王永国,郑君,等.基于三维激光扫描点云的树冠体积计算方法[J].农业机械学报,2013(7):235-240.
[11]樊仲谋,冯仲科,郑君,等.基于立方体格网法的树冠体积计算与预估模型建立[J].农业机械学报,2015(3):320-327.
[12]巩垠熙,何诚,冯仲科,等.基于改进Delaunay算法的树冠三维重构单木因子提取[J].农业机械学报,2013(2):192-199.
[13]徐伟恒,冯仲科,苏志芳,等.一种基于三维激光点云数据的单木树冠投影面积和树冠体积自动提取算法[J].光谱学与光谱分析,2014(2):465-471.
[14]张飞,谢步瀛,闫星宇,等.改进的三维点集凸包求取算法[J].计算机辅助工程,2009,18(1):78-82.
[15]黄俊杰,罗周全,秦亚光,等.复杂采空区三维散乱点建模技术研究及应用[J].东北大学学报(自然科学版),2016,37(12):1784-1788.
[16]吴文周,李利番,王结臣.平面点集凸包Graham算法的改进[J].测绘科学,2010(6):123-125.
[17]蒋联源.凸壳算法及其应用研究[D].南京:广西师范大学,2007.
[18]李必栋,闫浩文,王中辉,等.坐标排序的离散点凸包生成算法[J].测绘科学,2017,42(2):14-17.
[19]黄先锋,程晓光,张帆,等.基于边长比约束的离散点准确边界追踪算法[J].武汉大学学报(信息科学版),2009(6):688-691.
[2]林树森.城市道路交通长期规划应突出人居环境理念[J].城市与区域规划研究,2015(3):120-131.
[3]李杨,李秀峰.基于点云数据的树木骨架线提取研究[J].科技创新与生产力,2017(6):53-55.
[4]管西鹏.树木三维点云数据分析与建模技术研究[D].长沙:中南林业科技大学,2015.
[5]李宏星,欧阳玉华.基于三维激光扫描点云的树冠面积快速精准计算方法[J].绿色科技,2015(6):72-74.
[6]梁子瑜.基于TLS点云数据的林分调查因子测定及收获估计[D].南京:南京林业大学,2015.
[7]王祺,胡洪,吴艳兰,等.基于点云数据的树冠体积自动求算方法[J].西北林学院学报,2017,32(2):242-246.
[8]臧克.基于Riegl三维激光扫描仪扫描数据的初步研究[J].首都师范大学学报(自然科学版),2007,28(1):77-82.
[9]熊妮娜,王佳,罗旭,等.一种基于三维激光扫描系统测量树冠体积方法的研究——以油松为例[J].北京林业大学学报,2007(增2):61-65.
[10]韦雪花,王永国,郑君,等.基于三维激光扫描点云的树冠体积计算方法[J].农业机械学报,2013(7):235-240.
[11]樊仲谋,冯仲科,郑君,等.基于立方体格网法的树冠体积计算与预估模型建立[J].农业机械学报,2015(3):320-327.
[12]巩垠熙,何诚,冯仲科,等.基于改进Delaunay算法的树冠三维重构单木因子提取[J].农业机械学报,2013(2):192-199.
[13]徐伟恒,冯仲科,苏志芳,等.一种基于三维激光点云数据的单木树冠投影面积和树冠体积自动提取算法[J].光谱学与光谱分析,2014(2):465-471.
[14]张飞,谢步瀛,闫星宇,等.改进的三维点集凸包求取算法[J].计算机辅助工程,2009,18(1):78-82.
[15]黄俊杰,罗周全,秦亚光,等.复杂采空区三维散乱点建模技术研究及应用[J].东北大学学报(自然科学版),2016,37(12):1784-1788.
[16]吴文周,李利番,王结臣.平面点集凸包Graham算法的改进[J].测绘科学,2010(6):123-125.
[17]蒋联源.凸壳算法及其应用研究[D].南京:广西师范大学,2007.
[18]李必栋,闫浩文,王中辉,等.坐标排序的离散点凸包生成算法[J].测绘科学,2017,42(2):14-17.
[19]黄先锋,程晓光,张帆,等.基于边长比约束的离散点准确边界追踪算法[J].武汉大学学报(信息科学版),2009(6):688-691.