Next Article in Journal
Automatic Inundation Mapping Using Sentinel-2 Data Applicable to Both Camargue and Doñana Biosphere Reserves
Next Article in Special Issue
Assessment of Green Infrastructure in Riparian Zones Using Copernicus Programme
Previous Article in Journal
3D Coseismic Deformation Field and Source Parameters of the 2017 Iran-Iraq Mw7.3 Earthquake Inferred from DInSAR and MAI Measurements
Previous Article in Special Issue
Monitoring Green Infrastructure for Natural Water Retention Using Copernicus Global Land Products
Open AccessArticle

The Potential of Open Geodata for Automated Large-Scale Land Use and Land Cover Classification

German Remote Sensing Data Center, Earth Observation Center, German Aerospace Center, 82234 Wessling, Germany
Daimler Insurance Services GmbH, 70469 Stuttgart, Germany
Independent scholar, 80686 Munich, Germany
Author to whom correspondence should be addressed.
Remote Sens. 2019, 11(19), 2249;
Received: 23 August 2019 / Revised: 21 September 2019 / Accepted: 21 September 2019 / Published: 27 September 2019
This study examines the potential of open geodata sets and multitemporal Landsat satellite data as the basis for the automated generation of land use and land cover (LU/LC) information at large scales. In total, six openly available pan-European geodata sets, i.e., CORINE, Natura 2000, Riparian Zones, Urban Atlas, OpenStreetMap, and LUCAS in combination with about 1500 Landsat-7/8 scenes were used to generate land use and land cover information for three large-scale focus regions in Europe using the TimeTools processing framework. This fully automated preprocessing chain integrates data acquisition, radiometric, atmospheric and topographic correction, spectral–temporal feature extraction, as well as supervised classification based on a random forest classifier. In addition to the evaluation of the six different geodata sets and their combinations for automated training data generation, aspects such as spatial sampling strategies, inter and intraclass homogeneity of training data, as well as the effects of additional features, such as topography and texture metrics are evaluated. In particular, the CORINE data set showed, with up to 70% overall accuracy, high potential as a source for deriving dominant LU/LC information with minimal manual effort. The intraclass homogeneity within the training data set was of central relevance for improving the quality of the results. The high potential of the proposed approach was corroborated through a comparison with two similar LU/LC data sets, i.e., GlobeLand30 and the Copernicus High Resolution Layers. While similar accuracy levels could be observed for the latter, for the former, accuracy was considerable lower by about 12–24%. View Full-Text
Keywords: land use; land cover; classification; Copernicus; open geodata; CORINE; LUCAS; Landsat land use; land cover; classification; Copernicus; open geodata; CORINE; LUCAS; Landsat
Show Figures

Graphical abstract

MDPI and ACS Style

Leinenkugel, P.; Deck, R.; Huth, J.; Ottinger, M.; Mack, B. The Potential of Open Geodata for Automated Large-Scale Land Use and Land Cover Classification. Remote Sens. 2019, 11, 2249.

Show more citation formats Show less citations formats
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

Search more from Scilit
Back to TopTop