摘要
为了提高气象卫星资料的可用性，排除云干扰，本文根据云的时空分布变化特征，对可见光波段反射率和热红外波段温度进行分析和研究，提出了可见光反射率自动判云、热红外温度自动判云，可见光和热红外组合判云以及设立云区阈值判云等一系列检测方法。并对检测出的云区采用同周期相近时相的图像资料相对变化率来反演替代云区灰度值。保证了图像的连续性和客观性，取得了较好的效果。
Cloud removing is a key to expand application of meteorological satel-lite data.This paper analyses the cloud reflectance in visible band and temperaturein thermal infrared band from their time and spatial distribution.It also puts for-ward to a serial of detecting cloud methods,including automatic detection usingreflectance of visible band and using temperature of thermal infrared band,detec-tion using the combination of them and using the threshold of cloudy region.Forcloudy region,the pixel values are substituted by the image data of same cycle andnear time after being calculated according their correlation with data of cloudyimage.The images have better effect in continuity and reality through above proces-sing.
出处
《环境遥感》
CSCD
1995年第2期137-142,T001,共7页
关键词
云
气象卫星
检测法
Cloud,Automatic detection,Data substitution
