ESM_BUILT_VHR2018_Europe_R2020_3035_02 is a spatial raster dataset that is mapping human settlements in Europe, based on Copernicus Very High Resolution optical coverage for reference year 2018.
ESM 2018 exploits the Copernicus VHR_IMAGE_2018 dataset made of satellite images Pleiades 1A & 1B, SuperView-1, Kompsat-3/3A and PlanetScope with backup missions of Spot-6/7, TripleSat and Deimos-2 ranging from 2017 to 2019.
ESM_BUILT_VHR2018_Europe_R2020_3035_02 classifies the residential and non-residential buildings at a spatial resolution of 2 meters (EPSG:3035).
Corbane, Christina; Szabo, Filip (2026): ESM R2020 - European Settlement Map from Copernicus Very High Resolution data for reference year 2018. European Commission, Joint Research Centre [Dataset] doi: 10.2905/JRC.FWWAN4R PID: http://data.europa.eu/89h/192530fc-3d09-4740-8c1d-d1e601cdfdcc
building typologybuilt-up areasclassificationCopernicusESMGHSLLandPan-Europeanremote sensing
ESM_BUILT_VHR2018_Europe_R2020_3035_02 is a spatial raster dataset that is mapping human settlements in Europe, based on Copernicus Very High Resolution optical coverage for reference year 2018.
ESM 2018 exploits the Copernicus VHR_IMAGE_2018 dataset made of satellite images Pleiades 1A & 1B, SuperView-1, Kompsat-3/3A and PlanetScope with backup missions of Spot-6/7, TripleSat and Deimos-2 ranging from 2017 to 2019.
ESM_BUILT_VHR2018_Europe_R2020_3035_02 classifies the residential and non-residential buildings at a spatial resolution of 2 meters (EPSG:3035).
Project Web site
.svg){kind=icon}
.svg){kind=icon}
The ESM_2018 is the latest release of the European Settlement Map (ESM) and it is produced within the Global Human Settlement Project of the Joint Research Centre (JRC) following a request of DG REGIO. The ESM_2018 offers two raster products at 2m spatial resolution: i) The built-up area change detection layer and ii) Building footprints layer. Change detection layer is exploiting the pan-European Copernicus (Core_003) dataset for the reference year 2012 and Copernicus VHR_IMAGE_2018 dataset for reference year 2018. The later product is using only the VHR_IMAGE_2018 collection and it is developed for fine scale multiclass building delineation. Unlike its predecessors, the ESM_2015 and ESM_2012, the ESM_2018 workflow is based on the fully convolutional neural networks (instance segmentation). The ESM_2018 workflows were executed on Graphics Processing Unit (GPU) Docker containers on the JRC Big Data Analytics Platform (BDAP). Training data for the 2012-2018 change detection layer comes from Copernicus Urban Atlas Change 2012-2018 layer, while the training data for building detection layer was derived from Open Street Map and combined with Luisa 2018 basemap in order to discriminate between residential and non-residential buildings. Both products offer a discrimination between residential and non-residential buildings at 2m spatial resolution and at pan-European scale (EEA-39).
| From date | To date |
|---|---|
| 2018-01-01 | 2018-12-31 |
