Colored visualization of multitemporal SAR data for change detection: issues and methods

Konferenz: EUSAR 2018 - 12th European Conference on Synthetic Aperture Radar
04.06.2018 - 07.06.2018 in Aachen, Germany

Tagungsband: EUSAR 2018

Seiten: 4Sprache: EnglischTyp: PDF

Persönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt

Colin-Koeniguer, Elise (Onera, Chemin de la Hunière, 91123 PALAISEAU, France)
Boulch, Alexandre (ONERA, France)
Trouve-Peloux, Pauline; Janez, Fabrice (Onera, France)

This paper proposes a method to generate a colorful product for stacks of SAR images, making it easy to see areas that are subject to change. Two modes of representation are proposed: one intended for Sentinel changes, on a large temporal scale; the other for TerraSAR-X images, allowing you to see point-time events. The importance of fine coregistration and temporal calibration is emphasized. Lastly, filtering methods, in particular by deep learning, based on the temporal redundancy of the data, are proposed.