| 基于3D LiDar数据的城区植被识别研究 |
| |
| 引用本文： | 刘峰,龚健雅.基于3D LiDar数据的城区植被识别研究[J].地理与地理信息科学,2009,25(6). |
| |
| 作者姓名： | 刘峰 龚健雅 |
| |
| 作者单位： | 刘峰(中南林业科技大学理学院,湖南,长沙,410004;武汉大学测绘遥感信息工程国家重点实验室,湖北,武汉,430079);龚健雅(武汉大学测绘遥感信息工程国家重点实验室,湖北,武汉,430079) |
| |
| 基金项目： | 国家重点基础研究发展计划项目，中南林业科技大学青年科学基金项目，湖南省教育厅科研项目 |
| |
| 摘 要： | 提出一种基于支持向量机(SVM)的三维LiDar数据分类方法:利用kd-trees存储无序的点云数据,在局部邻域中利用点云数据间的几何关系估算植被表面特征值;将密度值和高程差值作为SVM输入特征变量,利用基于径向基函数的SVM方法实现植被点云数据的分类.实验结果为:OA分类精度达到94.31%,Kappa系数为89.53%.该方法操作性较强,在分类精度及计算效率方面比传统方法具有优势.
|
| 关 键 词： | LiDar数据 点云分析 支持向量机 城区植被 分类 |
Detection of Urban Vegetation with 3D LiDar |
| |
| LIU Feng,GONG Jian-ya.Detection of Urban Vegetation with 3D LiDar[J].Geography and Geo-Information Science,2009,25(6). |
| |
| Authors: | LIU Feng GONG Jian-ya |
| |
| Institution: | LIU Feng1,2,GONG Jian-ya2(1.School of Science,Central South University of Forestry , Technology,Changsha 410004,2.State Key Laboratory of Information Engineering in Surveying,Mapping , Remote Sensing,Wuhan University,Wuhan 430079,China) |
| |
| Abstract: | The Airborne Laser Scanner(ALS) can obtain high-resolution three-dimensional information about features of the earth surface.The three-dimensional laser scanning data have been used extensively in urban planning,urban ecological research and three-dimensional city modeling.Based on Support Vector Machine(SVM) of three-dimensional LiDar data classification methods,kd-trees was used to storage disorder of point cloud data,the different vegetation surface characteristic values in the local domain were achieved... |
| |
| Keywords: | LiDar data point cloud analysis SVM urban vegetation classification |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
