Abstract:
Earth observation can support the revealment of humanitarian crisis, especially if the location accessibility is reduced due to conflicts or governmental restrictions. An...Show MoreMetadata
Abstract:
Earth observation can support the revealment of humanitarian crisis, especially if the location accessibility is reduced due to conflicts or governmental restrictions. An important indicator is the change of artificial structures like buildings, roads and other sealed surfaces. In this paper, an AI-driven approach for a change analysis is shown. It focuses on the change of man-made structures leading to an increase or alteration in artificial coverage, which can be, in certain situations and considering additional contextual information, an evidence for humanitarian crisis. Beside a retrospective change detection, a hot-spot analysis is shown. The latter can serve as a starting point for further investigations using VHR satellite data. The use of deep learning techniques is superior to traditional change detection methods, especially when the object of change is selective and complex like artificial structures.
Date of Conference: 11-16 July 2021
Date Added to IEEE Xplore: 12 October 2021
ISBN Information: