| 基于支持向量机的遥感影像湿地信息提取研究* |
| |
| 引用本文： | 姚云军,张泽勋,秦其明,邱云峰.基于支持向量机的遥感影像湿地信息提取研究*[J].计算机应用研究,2008,25(4):989-990. |
| |
| 作者姓名： | 姚云军 张泽勋 秦其明 邱云峰 |
| |
| 作者单位： | 1. 北京大学,遥感与地理信息系统研究所,北京,100871
2. 北京大学,遥感与地理信息系统研究所,北京,100871;燕山大学,信息学院,河北,秦皇岛,066004 |
| |
| 基金项目： | 国家“863”计划资助项目(2006AA06Z233)，中国地质调查局资助项目(1212010510511) |
| |
| 摘 要： | 以ETM 影像数据为例,采用基于支持向量机的方法对黄河中上游区域湿地信息进行提取,并将该方法与传统的最大似然分类提取方法以及面向对象的提取方法进行对比分析。结果表明:基于支持向量机方法的提取精度高达93.57%,Kappa系数也超过了0.9,比单纯的最大似然分类方法或者面向对象的方法提取精度高得多,而且该方法操作性和实用性也很强。
|
| 关 键 词： | 支持向量机 ETM 影像 湿地信息 提取精度 |
| 文章编号： | 1001-3695(2008)04-0989-02 |
| 修稿时间： | 2007年3月27日 |
Study on wetland information extraction of remote sensing images based on support vector machine |
| |
| YAO Yun jun,ZHANG Ze xun,QIN Qi ming,QIU Yun feng.Study on wetland information extraction of remote sensing images based on support vector machine[J].Application Research of Computers,2008,25(4):989-990. |
| |
| Authors: | YAO Yun jun ZHANG Ze xun QIN Qi ming QIU Yun feng |
| |
| Affiliation: | (1.Institute of Remote Sensing & GIS, Peking University, Beijing 100871, China; 2.School of Information, Yanshan University, Qinhuangdao Hebei 066004, China) |
| |
| Abstract: | This paper analyzed the test-area ETM image data's features of Yellow River basin and utilized the method of wetland extraction based on support vector machine.In order to verify the extraction accuracy of SVM,carried out a comparison between this method and other methods,such as,the classification extraction method of maximum likehood,object-oriented approach.The case study shows that the accuracy of the method of SVM is 93.57% and Kappa index is above 0.9.The result from this method is more reasonable than the method of extraction based on maximum likehood or object-oriented approach.At the same time,this method can improve its operability and practicability. |
| |
| Keywords: | support vector machine(SVM) ETM images wetland information extraction accuracy |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《计算机应用研究》浏览原始摘要信息 |
|
点击此处可从《计算机应用研究》下载全文 |
|
