Local ontologies for object-based slum identification and classification

One of the consequences of rapid urbanization in developing countries is the proliferation of slums. Lack of quantified, updated and spatially disaggregated information on slums contributes to incomplete intervention and monitoring. The identification, delineation and characterization of slums spatially in a consistent manner will make it possible to target slum intervention programs as well as improve understanding of the phenomena geographically. We propose an ontological approach to define slums based on indicators related to the morphology of the built environment and address the issue of varying conditions by comparing two completely different contexts. The ontological indicators form the basis of object-oriented image analysis (OOA) and classification of slums. Examples from Kisumu (Kenya) and Ahmedabad (India) illustrate how these indicators might be applied.