AUTOMATED POPULATION AND DWELLING UNIT ESTIMATION FROM HIGH-RESOLUTION SATELLITE IMAGES - A GIS APPROACH

Methods were developed to extract population and dwelling unit data fr om a high-resolution multi-spectral SPOT image using an integrated dig ital image analysis and GIS approach for the Hong Kong metropolitan ar ea, which is characterized by high population densities and mixed land use located in multifunctional structures. These methods include diff erent forms of regression models linking spectral radiance values of i mage pixels with population densities, as well as with high and low re sidential densities. It was found that accurate estimates of populatio n and dwelling units could be obtained at the macro level for the whol e study area using any one of these models. However, at the micro leve l, the accuracy tended to be low because of the difficulty in discrimi nating residential use from non-residential use in multifunctional bui ldings in a satellite image.