Please use this identifier to cite or link to this item:
|Title:||Application of the Symbolic Machine Learning to Copernicus VHR Imagery: The European Settlement Map|
|Authors:||CORBAN CHRISTINA; SABO FILIP; SYRRIS VASILEIOS; KEMPER THOMAS; POLITIS PANAGIOTIS; PESARESI MARTINO; SOILLE PIERRE; OSÉ KENJI|
|Citation:||IEEE GEOSCIENCE AND REMOTE SENSING LETTERS vol. 17 no. 7 p. 1153-1157|
|Publisher:||IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC|
|Type:||Articles in periodicals and books|
|Abstract:||This letter introduces the new European Settlement Map (ESM) workflow, results and validation. Unlike the previous ESM versions, it uses the supervised learning combined with the textural and morphological features for built-up area extraction. Input data is the Copernicus very high resolution collection coming from a variety of sensors. The workflow is fully automated and it does not include any postprocessing. For the first time a new layer that classifies non-residential building is derived by using only remote sensing imagery and training data. The built-up area layer is delivered at 2m pixel resolution while the residential/non residential layer is delivered at 10m spatial resolution. More than 46000 scenes were processed and ~6 million km2 of Europe was mapped by using the Big Data infrastructure. Validation showed balanced accuracy of 0.81 and 0.91 for level 1 and 2 layers respectively and 0.70 for the non-residential layer.|
|JRC Directorate:||Space, Security and Migration|
Files in This Item:
There are no files associated with this item.
Items in repository are protected by copyright, with all rights reserved, unless otherwise indicated.