Remote sensing of aerosols from satellite is the most convenient tool for providing global aerosol spatial and temporal distributions. It is difficult to discriminate the aerosol signal from the surface reflectance, that is the reason why the algorithms based only on visible spectral region fail particularly over bright surfaces. Our algorithm developed to detect aerosol contaminated-pixels uses both MSG-SEVIRI infrared and visible observations. Aerosol microphysical parameters are retrieved for pixels affected by aerosol in an iterative manner by simultaneously comparing satellite radiance observed at visible, near-infrared and infrared wavelengths to look-up tables of simulated radiances.