摘要
湿地资源遥感信息提取并满足一定的分类精度和效率,是湿地研究的关键问题之一。利用高分辨率的SPOT5影像数据,通过对研究区分别进行基于像素的监督分类和基于对象的分类,进行典型沼泽湿地专题信息提取,同时将面向对象分类与DEM辅助数据进行信息复合,探寻提升分类精度的方法。结果表明,基于对象和辅助数据DEM的信息提取,可以有效地改善遥感图像的分类精度。
Remote sensing information extraction of wetland resource is one of the most important aspects in wetland research that need to meet definite accuracy and efficiency. Based on SPOT5 image, thematic information extraction of typical wetland was studied by carrying out pixel-based supervised classification and object-based classification in study area,and DEM data was combined with the information of object-oriented classification to explore ways to improve the accuracy of classification. The results showed that the information extraction based on objects and DEM can effectively improve the accuracy of remote sensing image classification.
出处
《遥感技术与应用》
CSCD
北大核心
2009年第3期346-351,I0005,共7页
Remote Sensing Technology and Application
基金
国家863项目"湿地复杂地表环境信息自动提取及生境退化特征定量分析评价技术示范研究"(2007AA12Z176)资助
关键词
洪河湿地
信息提取
面向对象
DEM
Honghe wetland
Information extraction
Object-oriented
DEM
