Detection of Urban Buildings with Use of Multispectral Gokturk-2 And Sentinel 1A SAR Images

Urban areas are important for city planning, security, traffic purposes, an decision makers etc. Remotely sensed data are useful to detect the urban areas either with active or passive systems. Each system has advantages and disadvantages. Passive images are mainly multispectral images, and they have rich information with their rich spectral resolution. On the other hand, they are effected with atmospherical conditions, so there should not be clouds over the sensed region during the data acquisition. On the other hand, SAR systems are not affected by the atmospheric conditions, but their spectral resolution is low, mainly with one channel. On the other hand, the structure of the passive images is completely different than the multispectral images. Secondly, the geometrical and electrical properties of objects play an important role in the pixel values. In this study, multispectral GOKTURK-2 MS image and SENTINEL 1A SAR image have been used to detect the urban buildings to use the advantages of the both datasets. Firstly, SVM method is applied to detect the buildings in GOKTURK image. The buildings are detected from SAR image with fuzzy logic approach. Finally, the buildings have been detected with intersection of the both results. The results from SAR image could eliminate false negative results from GOKTURK-2 image. Study area is selected from Antalya, Kepez distinct. The detected urban area is 288.353 m² at the selected study area.