Automatic Vehicle and Occlusion Removal in Aerial and Satellite Imagery Using Deep Learning Algorithms

Kurzfassung

Occlusion-free drivable areas in aerial and satellite imagery can be required in various applications, such as traffic control, city modeling and navigation systems. However, these areas, including roads and parking places, are typically highly occluded by vehicles, buildings, bridges, pedestrians, trees, and some other roads. Although the latest deep learning algorithms have made significant contributions to image inpainting for object removal, there is still no appropriate resolution to specific tasks such as automatic removal of occlusions in images which were taken aerially. This thesis attempts to investigate an effective approach, which can not only eliminate occlusions, but also restore interesting features of target areas. In addition to pure road surface and parking places, other characteristics such as lane-markings and curbstones are also viewed as reconstructed features since they are used by vehicles in general transportation infrastructure.