Rule-based impervious surface mapping using high spatial resolution imagery

Impervious surface mapping has become a recent concern in remote-sensing applications because of worldwide urban growth and the resultant environmental changes. However, many effective impervious surface mapping techniques developed for moderate-resolution imagery are not applicable to high spatial resolution imagery such as that of IKONOS and the Advanced Land Observing Satellite (ALOS) due to their limited number of spectral bands and the lack of middle-infrared bands. In response to this need, this article proposes a rule-based approach to extract impervious surface features from high spatial resolution imagery. Images of IKONOS, ALOS and Systeme Pour l'Observation de la Terre 5 (SPOT-5) from three different cities were used for the approach. The general rules used for extracting impervious surface features from the images were deduced from the signatures of land cover classes but are not scene-specific. Based on these rules, impervious surface features were effectively discriminated from the categories such as soil and water, which were always confused with each other in previous work. The approach can achieve a high overall accuracy in imperviousness extraction efforts.