首页 | 本学科首页   官方微博 | 高级检索  
     

决策树在遥感影像分类中的应用
引用本文:巴桑,刘志红,张正健,李奇穗. 决策树在遥感影像分类中的应用[J]. 四川气象, 2011, 0(2): 31-34
作者姓名:巴桑  刘志红  张正健  李奇穗
作者单位:西藏高原大气环境科学研究所;成都信息工程学院;四川省眉山市气象局;
基金项目:国家发改委“十一五”重点项目“西藏自治区生态环境遥感监测与服务系统”
摘    要:在充分了解决策二叉树分类过程的基础上,以西藏波密地区的TM遥感影像为研究对象,提取并分析各种典型地物的反射光谱特征,利用不同地物的光谱差异并结合人工目视判读建立了一组能够精确、快速划分遥感影像的分类规则。用ENVI建立分类决策树对实验区遥感影像进行分类,并将分类结果和传统的最大似然法相比较。结果表明:决策树分类的总体精度和Kappa分别为97.15%和0.9644,优于传统的最大似然法,取得了较为满意的分类结果。

关 键 词:决策树  遥感影像  最大似然法  分类  误差矩阵

Decision Tree and Its Application in Remote Sensing Image Classification
BA Sang,LIU Zhihong,ZHANG Zhengjian,LI Qisui. Decision Tree and Its Application in Remote Sensing Image Classification[J]. Journal of Sichuan Meteorology, 2011, 0(2): 31-34
Authors:BA Sang  LIU Zhihong  ZHANG Zhengjian  LI Qisui
Affiliation:BA Sang1,LIU Zhihong2,ZHANG Zhengjian2,LI Qisui2,3(1.Tibet Institute of Plateau Atmospheric and Environmcntal Sciences,Lasa 850001,2.Chengdu University of Information Technology,Chengdu,610225,3.Meteorological Bureau of Meishan County,Meishan,620010)
Abstract:Taking TM images in Bomi county of Tibet as the object to study,choosing and analyzing the reflect spectral characteristics of typical land-cover types,a set of accurately and fast rules for TM images classification has been established based on spectral characteristics diversity of different land-cover types and artificial visual interpretation.The classification decision tree was applied to classify the TM images in experimental area,and the classification results were compared with traditional maximum li...
Keywords:decision tree  remote sensing image  maximum likelihood  classification  error matrix  
本文献已被 CNKI 维普 等数据库收录!
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