Inference of top of the atmosphere clear{sky re-ectances associated to meteorological satellite products are crucial prior to estimate cloud parameters such as optical depth and phase, clouds playing a major part in the Earth radiation budget. In this poster, we present an innovative method to estimate composite TOA clear{sky re-ectances from visible geostationary satellite measurements which is based on a statistical approach of time{series re-ectance values by use of the quantile corrected in order to account for the scene geometry dependence of each pixel. The bene ts and simplicity of this new approach is discussed towards some major algorithms found in the literature. It is then applied to 7 months of Meteosat{7 daytime visible observations. This algorithm is planned to be implemented in the Spinning Enhanced Visible and InfraRed Imager (SEVIRI) scene identi cation which is a part of the Geostationary Earth Radiation Budget (GERB) Ground Segment developed at RMIB.