Urban building extraction through object-based image classification assisted by digital surface model and zoning map

This study develops an object-based image classification methodology for urban land covers classification, using very high resolution aerial images, elevation data and city zoning maps. Logically structured classification rules based on spectral, spatial and contextual features of the segmented objects are first created and tested over a small urban area. The same rule set is then transferred and tested on two similar images covering larger urban areas. The land cover classification results through the transferability of the rule set prove the effectiveness of the methodology and produce satisfactory classification results with an overall accuracy of 91% as against 96% that was achieved over the small representative training area. The classification methodology based on the integrated use of multiple data produces satisfactory land cover classification. Its transferability considerably reduces both the processing time and the analyst's efforts.