首页 | 本学科首页   官方微博 | 高级检索  
     检索      

利用面向对象分类技术的大尺度土地覆被调查方法
作者姓名:罗开盛  李仁东  常变蓉
作者单位:1. 中国科学院测量与地球物理研究所, 武汉 430077; 2. 中国科学院大学, 北京 100049; 3. 湖北省环境与灾害监测评估重点实验室, 武汉 430077
基金项目:中国科学院战略性先导科技专项(XDA0505107)资助
摘    要:以HJ-CCD为实验数据,采用面向对象分类技术,对地形复杂、类型多样的湖南省进行土地覆盖类型的自动提取.着重研究在大尺度上的土地覆被调查中应用HJA/B遥感影像和面向对象技术获取土地覆被信息的一整套技术方法.将多尺度分割、邻域推移分类法以及野外调查、专家知识有机结合起来,并用野外采样点进行精度检验.湖南省土地覆被调查结果的总体精度84.99%,Kappa系数为82.79%.结果证明了以HJ-CCD影像为遥感数据源,利用面向对象技术进行大尺度的土地覆被调查的可行性和有效性.

关 键 词:HJ-CCD影像    面向对象    大尺度    邻域推移分类法    ecognition
收稿时间:2013-03-01
修稿时间:2013-03-20

Land-cover survey method using object-oriented technology and HJ-CCD image on large scale
Authors:LUO Kai-Sheng  LI Ren-Dong  CHANG Bian-Rong
Institution:1. Institute of Geodesy and Geophysics, Chinese Academy of Sciences, Wuhan 430077, China; 2. University of Chinese Academy Sciences, Beijing 100049, China; 3. Key Laboratory for Environment and Disaster Monitoring and Evaluation of Hubei Province, Wuhan 430077, China
Abstract:Based on HJ-CCD remote image data, object-oriented technology was applied into the land-cover extraction in Hunan Province. We mainly focus on land-cover survey method using HJA/B remote sensing images on large scale and explore a set of approaches to get land-cover information using object-oriented technology. The overall accuracy of the results is 84.99% and Kappa coefficient is 82.79%. The research results show that land-cover survey method using HJ-CCD remote sensing images and object-oriented classification technology is feasible and effective on large scale.
Keywords:HJ remote sensing image                                                                                                                        object-oriented                                                                                                                        large scale                                                                                                                        partition mobile classification method                                                                                                                        ecognition
点击此处可从《》浏览原始摘要信息
点击此处可从《》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号