URBAN LAND USE MAPPING BASED ON OBJECT-BASED IMAGE ANALYSIS USING WORLDVIEW-3 SATELLITE IMAGERY

摘要

Land Use and Land Cover (LULC) data are important to monitor and assess environmental change. LULC classification using satellite images is a method widely used on a global and local scale especially, in urban areas that have various land cover types that are important components of the urban landscape and ecosystem. The objectives of this study aim to classify urban land use using WorldView-3 using very high spatial resolution satellite imagery and the Object-Based Image Analysis (OBIA) method. A decision rule set was applied to classify the WorldView-3 images in Kathu sub-district, Phuket province, Thailand. The main steps were as follows: (1) the image was orthorectified with ground control points and using the Digital Elevation Model (DEM), (2) multi-scale image segmentation was applied to divide the pixel image to image object level, (3) development of the rules set for LULC classification using spectral bands, spectral indices, spatial and contextual information and (4) accuracy assessment was made using testing data which sampled using statistical random sampling. The results show that six classes (vegetation, grass, water, road, open space, built-up area) were been successfully classified with overall classification accuracy of 87.61% and a Kappa coefficient of 0.84. In terms of producer's accuracy, we achieved more than 85% accuracy for water, vegetation, grass and built-up area. Open space and road is still fairly difficult to identify.