Abstract
Information about the spatial and structural properties as well as different indicators of social and economic functions cannot be easily extracted from remote-sensing data in an urban milieu. This paper focuses on the extraction of information that is relevant to the strategic spatial level of urban planning management, i.e., more general land-use descriptions, using the window-independent context segmentation method to extract urban area categories from a SPOT4 satellite scene. In this study, we were able to extract three different urban categories, industrial/commercial, and two residential categories that belong to different suburbanisation phases.
Acknowledgements
This work was supported by the Sweden-America Foundation and the Riksbankens Jubileumsfond. The SPOT4 image was acquired through the SPOT data/ISIS Programme, CNES Copyright.