NOAA卫星云雾自动检测和修复方法  

Automatic detection and rehabilitation of cloud / fog pollution in NOAA/AVHRR imagery

作　　者：周红妹[1] 谈建国[1] 葛伟强[1] 杨崇俊[2] 

机构地区：[1]上海市气象科学研究所,上海200030 [2]中国科学院遥感应用研究所,北京100101

出　　处：《自然灾害学报》2003年第3期41-47,共7页Journal of Natural Disasters

基　　金：国家高新技术"八六三"项目(818-07-01)

摘　　要：云雾在时间和空间上具有不稳定性,不同高度、厚度、面积和太阳高度角情况下,NOAA卫星可见光波段的海面反射率和红外波段亮度温度都不相同。据此,根据云雾光谱和红外特征,研究并找出了可见光波段海面最小反射率、云雾区最大反射率与云雾最佳判别阈值之间的关系,得出了不同月份东海海面最低温度和NOAA卫星红外波段亮温特征;通过采用光谱和结构分析相结合的方法,研究和研制了NOAA气象卫星云雾自动检测和修复模型,取得了较好的运行和检测、修复效果。Cloud and fog behave as unstability in time and space. So there are some differences of reflectivity in the visible region and of brightness temperature in the infrared region between non cloud/fog areas and cloud/fog areas in condition of its different solar zenith, different altitude, different area and thickness in different season. Upon these the relative relationship between the minimum reflectivity of ocean surface, maximum reflectivity of cloud/ fog, and the optimal cloud/fog-distinguishing threshold is found. Furthermore, the characteristic of minimum SST and brightness temperature in infrared wave band are obtained through NOAA/AVHRR data. By combinational spectrum and structure analysis, the model of automatic cloud/fog detection and rehabilitation of NOAA/AVHRR imagery is established and the efficient operational result and the good precision are acquired.

关 键 词：NOAA卫星 云雾 自动检测 修复 光谱特征 结构特征 

分 类 号：P412.27[天文地球—大气科学及气象学]