摘要
利用Landsat-5TM,Landsat-7ETM+和Landsat-8OLI遥感影像数据和高空间分辨率(1.14,0.54m)的Google Earth^(TM)地面实况图进行水库面积检测与变化监测的技术研究.采用完全约束最小二乘法混合像元分解(FCLS)方法提取不同水库水体信息,并与常用水体指数法及支持向量机(SVM)分类法进行比较,利用多时相FCLS方法监测厦门杏林湾水库2006-2014年间的面积动态变化.结果表明:基于FCLS的方法比NDWI等水体指数方法更能准确地获取水库水体面积大小及其变化信息.
We use Landsat-5TM ,Landsat-7 ETM＋,and Landsat-8 OLI remote sensing images and high spatial resolution （1.14,0.54m） Google EarthTM ground truth images to extract surface water and detect changes of reservoir area. Using fully constrained least squares method pixel unmixing （FCLS） method to ex-tracting different reservoir surface water information and compared with water index method and support vec-tor machine （SVM）. Finally, using multi-temporal FCLS method to dynamic monitoring Xinglin Bay Reservoir water area in Xiamen between 2006 and 2014. The experimental results showed that： compared with NDWI and other indexes method, FCLS based method can obtain more accurate dynamic information of the reservoir sur-face water area.
作者
孔美美
陈锻生
KONG Meimei CHEN Duansheng(College of Computer Science and Technology, Huaqiao University, Xiamen 361021 , China)
出处
《华侨大学学报（自然科学版）》
北大核心
2017年第3期385-390,共6页
Journal of Huaqiao University(Natural Science)
基金
国家自然科学基金面上资助项目(61370006)
福建省科技计划(工业引导性)重点项目(2015H0025)
关键词
水体面积提取
混合像元分解
遥感
变化检测
环境监测
杏林湾水库
water area extraction
pixel unmixing
remote sensing
change detection
environment monito-ring
Xinglin Bay Reservoir
