摘要
以南京市为研究区,基于面向对象的分类方法,对TM影像进行LBV变换,选择B分量阈值进行水体信息提取.结果表明,水体信息提取精度为94.6%.与基于像元的B分量阈值法和基于像元及面向对象的改进的归一化水体指数(MNDWI)阈值法相比,该方法提取水体信息快速准确,不但消除了山体阴影和城市建筑物的影响,还减少了分类"椒盐现象".该方法简单易行,适用于水体信息的自动提取.
Taking the Nanjing city as the research area,we transform the TM image with the LBV way to extract the water information based on object-oriented classification method.The experiment results show that the extraction accuracy of water information reaches 94.6%.This method not only removes the influence of the mountain shadow and city buildings,but also reduces the salt-phenomenon.This method is simple and has practical value.It is suitable for automatic extraction of water information.
出处
《中国科学院研究生院学报》
CAS
CSCD
北大核心
2012年第6期775-779,共5页
Journal of the Graduate School of the Chinese Academy of Sciences
基金
中国科学院战略性先导科技专项(XDA05050106)资助
关键词
LBV变换
面向对象
多尺度分割
水体提取
MNDWI
LBV transformation
object-oriented method
multi-segmentation
extraction of water
MNDWI
