The goal of this diploma thesis is the development of a method for the actualization of existing land cover datasets using current satellite images. For this, the kNN (k nearest neighbours) method was used as a classification method. The satellite images are classified using reference data from outdated land cover maps. The classification procedure is able to detect inconsistencies between the spectral characteristic of the satellite images and the reference data and can thereby eliminate those portions of the reference data, which are incorrect or out of date. This method was applied to the land cover datasets SINUS and CORINE and shows good results. As a practical example, a land cover map of Lower Austria with 13 land cover categories was generated.