Cloud cover influences the distribution of the solar radiation reaching the ground. Resulting fluxes of sensible and latent heat are critical to the accurate characterization of boundary layer behavior and mesoscale circulation that often lead to convective developments. Therefore the spatial and temporal variation in cloud cover can greatly affect regional and localized weather processes. This work describes an algorithm aimed at identifying cloudy and clear pixels in MODIS images. The algorithm is a real time procedure which, from geolocated and calibrated data, allows obtaining cloud masks with four clear sky confidence levels.