<
doi:  10.12013/qxyjzyj2019-002
基于光谱分析和综合阈值的FY-3/MERSI云检测

Cloud detection based on spectrum analysis and synthetical threshold for FY-3/MERSI data
摘要点击 132  全文点击 70  投稿时间:2019-02-01  修订日期:2019-02-19
查看HTML全文  查看全文  查看/发表评论  下载PDF阅读器
基金:  江西省重点研发计划项目(编号:20161BBG70093); 江西省科技计划项目(编号:2015ZBBG70041).
中文关键词:  MERSI,云检测,光谱分析,云检测指数
英文关键词:  MERSI  cloud detection  spectrum analysis  normalized cloud detection index
           
作者中文名作者英文名单位
樊建勇Fan Jianyong江西省气象科学研究所
陈兴鹃Chen Xingjuan江西省生态气象中心
聂志强Nie Zhiqiang江西省生态气象中心
李翔翔Li Xiangxiang江西省农业气象中心
引用:樊建勇,陈兴鹃,聂志强,李翔翔.2019,基于光谱分析和综合阈值的FY-3/MERSI云检测[J].气象与减灾研究,42(1):9-15
中文摘要:
      云检测是卫星遥感影像应用中的重要环节。为了获取FY-3中分辨率光谱成像仪(Medium Resolution Spectral Imager,MERSI)传感器精确的云掩膜数据,在分析云与植被、水体、裸露地表等地物在不同波段光谱差异的基础上,结合FY 3/MERSI数据的通道特性,构建基于多光谱和综合阈值的云检测算法。采用热红外通道亮温(CH5)检测出高云,使用云检测指数(Normalized Cloud Detection Index,INCD)和可见光波段反射率(CH3)检测中低云。利用江西地区的FY-3/MERSI数据进行应用,将检测结果与其他算法、人机交互解译结果进行了对比分析。结果表明,该算法对FY-3/MERSI影像上的云区具有较好的检测效果,同时利用地面气象观测资料进行验证,检测精度均高于88%,能够满足当前云检测精度需求。
Abstract:
      Cloud detection is vital for the application of satellite data. Based on the difference of spectrum in different bands between cloud and vegetation, water, exposed surface , etc., combined with the channel characteristics of the Medium Resolution Spectral Imager (MERSI) instrument, a multispectral cloud detection algorithm using MERSI images was presented, in order to identify cloud pixel accurately from MERSI sensor. High cloud was detected by using the brightness temperature (CH5). Low medium cloud was detected by using Normalized Cloud Detection Index and the reflectance (CH3). The distribution of cloud in Jiangxi Province was detected using the algorithm, and the results were validated by comparing to other algorithms result and the actual cloud image. The analysis indicated that the algorithm can accurately detect the distribution of cloud using the FY-3/MERSI images. Moreover, the surface meteorological observation records were also used to verify the cloud detection result and its accuracy was higher than 88%.
主办单位:江西省气象学会 单位地址:南昌市高新开发区艾溪湖二路323号
联系电话: (0791)82713175传真:0791-82713175 邮编:330096 Email:qxyjzyj@163.com
本系统由北京勤云科技发展有限公司设计