Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《Science of Surveying and Mapping》 2017-01
Add to Favorite Get Latest Update

The support vector machine method for RS images' classification in northwest arid area

ZHANG Jing;ZHAG Xiang;TIAN Long;ZHANG Qingfeng;College of Natural Resources &Environment,Northwest A&F University;  
Aiming at the classification method for the field of land resources of remote sensing images,which cover a large area with different climate and geomorphology at different phases,this paper provides a support vector machine(SVM)method based on the NDVI and texture feature.It improve the classification precision of the northwest arid areas at Yanan in Shaanxi,Jiayuguan in Gansu and Guoluo in Qinghai,solved the problem of not high classification accuracy in the maximum likelihood and BP neural network method because of their own defects.The result shows that:compared with the maximum likelihood method and BP neural network method,the SVM method has the highest classification accuracy(97.75%)with its Kappa coefficient was 0.9691.The SVM method can provide a methodological reference for the land resources sustainable development strategic.
【Fund】: 国家科技支撑计划子课题项目(2011BAD29B09-1-1C);; 中央高校基本科研业务费专项资金项目(QN2012042);; 西北农林科技大学博士科研启动基金项目(201104050395)
【CateGory Index】: P237
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved