摘要
基于ETM+遥感卫星数据,使用层次分析法、光谱响应特征阈值算法、监督分类和无监分类方法,对实验区土地利用信息进行提取。研究结果表明:采用层次分析法对ETM+数据进行土地利用调查,可最大程度识别出土地利用类型;综合利用多种信息提取方法进行山区土地利用与土地覆盖信息提取具有速度快、技术操作简单、可识别类型多的特点;对林地和水域的分类识别精度可达85%以上,但对山区农村居民地的面积难以准确提取,必须借助人工解译方法,并参考其他辅助资料才能获取。
This paper describes the process on using the methods combined spectral threshold algorithm, supervised classification, unsupervised classification and manual interpretation based on Landsat ETM＋ to derive Land-use and Land-cover information. The results show that the scheme of multi-stage classification combined different information extracting techniques work well for most cover-types. The accuracies for identifying forest land and water area are more than 85%. But it is difficulty to derive some cover-types correctly, such as small resident area and small road in the mountains.
出处
《云南地理环境研究》
2005年第6期34-37,共4页
Yunnan Geographic Environment Research
基金
国家自然科学基金资助项目(40401022)
关键词
土地利用
土地覆盖
遥感
信息提取
land-use
land-cover
remote sensing
information extraction
