首页 | 本学科首页   官方微博 | 高级检索  
     检索      

FY3B/MERSI数据的湖泊湿地面积提取对比分析
引用本文:冯锐,于文颖,纪瑞鹏,张玉书,武晋雯,陈鹏狮.FY3B/MERSI数据的湖泊湿地面积提取对比分析[J].测绘科学,2017,42(7).
作者姓名:冯锐  于文颖  纪瑞鹏  张玉书  武晋雯  陈鹏狮
作者单位:1. 中国气象局沈阳大气环境研究所,沈阳,110166;2. 辽宁省气象科学研究所,沈阳,110166
摘    要:针对国产风云三系列中分辨率卫星快速有效地进行水体识别的问题,基于2011—2016年数据进行了辽宁省主要湖泊水库的光谱分析,提出了晴空条件和有云情况下分别采用归一化水体指数方法和通道值与归一化水体指数相结合方法进行湖泊水库的提取。结果表明:FY3B/MERSI湖泊光谱曲线具有水体光谱特征,具体表现为8通道的数据反射率最高,18通道数据反射率最低。通过与49景TM数据水体提取结果进行对比,FY3B/MERSI数据水体提取的面积精度达到85%以上,总体分类精度达到90%以上,Kappa系数在0.56~0.95之间。提出了简便、快捷的计算模型,为国产卫星数据的业务应用提供了初步方法。

关 键 词:MERSI数据  湖泊水库  面积提取  方法比较

Comparison and analysis of several extraction methods for lakes and reservoirs based on FY3B/MERSI
FENG Rui,YU Wenying,JI Ruipeng,ZHANG Yushu,WU Jinwen,CHEN Pengshi.Comparison and analysis of several extraction methods for lakes and reservoirs based on FY3B/MERSI[J].Science of Surveying and Mapping,2017,42(7).
Authors:FENG Rui  YU Wenying  JI Ruipeng  ZHANG Yushu  WU Jinwen  CHEN Pengshi
Abstract:Aiming at the problem that of fast and effective water identification based on domestic satellite,the method were proposed that the NDWI based model showed higher precision under the clear sky,while the single band method and NDWI based model showed higher precision under cloudy sky.Using the data of FY3B/MERSI from 2011 to 2016,the characteristics of the main lakes and reservoirs reflective spectrum were analyzed in liaoning province,Meanwhile,the water extraction method were established by comparing the usability of the method,such as the single band method,the normalized difference water index(NDWI),the vegetation index method(RVI),the normalized difference vegetation index method (NDVI),and the models were tested by extracting water body from TM image with supervised classification method.The results show that the spectral reflectance are in accordance with the characteristics of water in visible band and near-infrared band,The peak appears at the NO.8 channel and the valley appears at the NO.18 channel of the FY3B/MERSI.The extracting results were compared with TM images,area extraction accuracy reached 85%,the overall classification accuracy are achieved 90% or more,the kappa coefficient between 0.56 to 0.95.
Keywords:MERSI data  lakes and reservoirs  area extraction  comparison of methods
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号