A lot of our work uses satellite images that need to go through a series of processes before they can be in a shape that can be further analyzed to create urban data. In this case, we use a deep-unfolded method to remove clouds and their shadow obstructions from the images. Clouds, together with their shadows, usually occlude ground-cover features in optical remote-sensing images. This hinders the utilization of these images for a range of applications such as earth observation, land-cover classification, and other urban data uses.
 We propose a deep unfolded and prior-aided robust principal component analysis (DUPA-RPCA) network for removing clouds and recovering ground-cover information in multitemporal satellite images. This is the first deep-unfolded method for cloud removal and has been published in IEEE Signal Processing Letters.