The aim of the paper is to present automated methods of discrimination of urban development forms using object-oriented classification in high-resolution images taken by the
The research covered an area of 25 km, situated in central Warsaw. As a result of object-oriented classification, five classes of development typical of large cities were distinguished and described.