 This paper presents a methodology for producing accurate land cover maps at country scale using high-resolution optical image time series, which is based on supervised classification and uses existing databases as reference data for training and validation. The approach uses all available image data, a simple preprocessing step leading to a homogeneous set of acquisition dates over the whole area, and a robust supervised classifier. The produced maps have a kappa coefficient of 0.86 with 17 land cover classes and are provided with a confidence map giving information at the pixel level about the expected quality of the result. The processing is efficient, does not need expensive field surveys or human operators for decision making, and uses open and freely available imagery. This article was authored by Jordi Inglata, Arthur Vincent, Marcella Arias, and others.