Improving Urban Land Cover Classification with Combined Use of Sentinel-2 and Sentinel-1 Imagery

Accurate land cover mapping is important for urban planning and management. Remote sensing data have been widely applied mapping. However, obtaining classification via optical remote alone difficult due to spectral confusion. To reduce the confusion between dark impervious surface water, Sentinel-1A Synthetic Aperture Rader (SAR) are synergistically combined with Sentinel-2B Multispectral Instrument (MSI) data. The novel support vector machine composite kernels (SVM-CK) approach, which can exploit spatial information, proposed process combination of MSI SAR based on fusion yields an overall accuracy (OA) 92.12% a kappa coefficient (KA) 0.89, superior results using imagery separately. indicate that inclusion improve performance by reducing built-up area water. This study shows be improved fusing imagery.