摘要
针对目前遥感水系提取存在的问题,提出了一种基于数学形态学的遥感影像水系提取方法。该方法在光谱模式识别提取水体信息的基础上,利用数学形态学方法将提取的水体结果进行断线连接、去噪及细化等处理,从而得到连续的水系。以Landsat ETM+影像为例进行了实验,结果表明,该方法可以提取连续、完整的水系;且提取的水系精度与现势性均优于国家测绘局发布的1∶25万数字线划图(DLG)水系。
A mathematical morphology based drainage line extraction method from remote sensing image is proposed.The coarse drainage line is first extracted with spectral pattern recognition approach,and then is subject to processes such as drainage line connection,noise removal and drainage line thinning using mathematical morphology transformations.A subset of landsat ETM＋ image is used to test the presented method.The result shows that this method can effectively extract continuous and smooth drainage lines.Further comparison is made between the extracted drainage lines and that of the digital line graph（DLG） with reference of remote sensing image.It demonstrated that the drainage lines by the proposed method are better agreed with the image than that of DLG.
出处
《武汉大学学报（信息科学版）》
EI
CSCD
北大核心
2011年第8期956-959,共4页
Geomatics and Information Science of Wuhan University
基金
国家自然科学基金主任基金资助项目(41040021)
湖北省自然科学基金资助项目(2009CDB104)
中央高校基本科研业务费专项资金资助项目(CUGL100211)
武汉大学测绘遥感信息工程国家重点实验室开放研究基金资助项目((09)重01)
关键词
水系
数学形态学
遥感
数字线划图
drainage line
mathematical morphology
remote sensing
digital line graph
