A Normalization Procedure for METEOSAT Visible Channel Data

Abstract

An efficient normalization scheme for satellite visible channel data is presented. The method is based on the percentile hypothesis. It is applicable to all weather scenes without any need for scene classification and detection techniques. Performance assessment with METEOSAT-2 visible data from March to October 1987, subsequent to parameter adjustment with May to August 1987 data, shows that the procedure, which accounts for the solar zenith angle, produces reasonable results for solar zenith distances up to almost 90°. The normalization reference angle may be chosen in order to optimize the brightness range of the normalized picture. Simple theoretical considerations allow insight into the structure of the scheme.