Mapping CORINE Land Cover from Sentinel-1A SAR and SRTM Digital Elevation Model Data using Random Forests

摘要: The European CORINE land cover mapping scheme is a standardized classification system with 44 and use classes. It used by the Environment Agency to report large-scale change minimum unit of 5 ha every six years operationally mapped its member states. most commonly applied method map visual interpretation optical/near-infrared satellite imagery. Sentinel-1A carries C-band Synthetic Aperture Radar (SAR) was launched in 2014 Space as first operational Copernicus mission. This study investigation for mapping. Two images acquired during ramp-up phase May December over Thuringia Germany are analysed. 27 hybrid level 2/3 classes defined. 17 these were present at site classified based on stratified random sample training pixels from polygon-eroded 2006 map. logarithmic radar backscatter HH HV polarisation (May acquisition), VV VH (December image texture input bands classification. In addition, Digital Terrain Model (DTM), Canopy Height (CHM) slope aspect maps Shuttle Topography Mission (SRTM) account geomorphological features landscape. future, elevation data will be delivered areas sufficiently high coherence Interferometric Wide-Swath Mode itself. When augmented interferometry, able discriminate several classes, making it useful monitoring cloud-covered regions. A bistatic Sentinel-1 Convoy mission would enable single-pass interferometric acquisitions without temporal decorrelation.