Land Use Classification

Exploitation of Copernicus data for spatial observation (incora)

Challenge

The classifications have a spatial resolution of 10m x 10m (pixel size) and represent the classes of built-up area, grassland, forest, agriculture, vegetationless soil and water. Atmospheric-corrected Sentinel-2 data provided by the EOC Geoservice of the German Aerospace Center (DLR) are used as the data basis with the MAJA algorithm. The training data required for the classification are generated rule-based from reference data (e.g. OpenStreetMap) and the Sentinel-2 scenes themselves. This allows processing to be flexibly transferred to different years and carried out fully automatically in the cloud. Information needs and indicator concepts were identified and coordinated between partners from science, administration and planning practice.

Services

The company mundialis implemented the following work steps in the incora project:

  • Development of an indicator subsystem to qualify settlement area monitoring from Copernicus products.
  • Development of processing rules for the evaluation of remote sensing data
  • Iterative implementation of indicator candidates (scenes, area-wide, time series).
  • Processing nationwide and in selected test and reference regions
  • Evaluation of the results with recognized benchmark methods (classification quality, duration, costs, etc.)
  • Provision of indicators for CODE-DE services
  • Reporting and scientific publications
  • Development of a WebGIS platform and integration of the results into existing platforms for the publication of project results
  • Provision of time series for the modeling of future settlement area development

The result classification consists of the following land cover classes:

  • Forest
  • Low vegetation
  • Water
  • Structural area
  • Area without vegetation
  • Agriculture
Result
  • Development of process chains for the automated production of land cover maps based on Copernicus satellite data.
  • Preparation of the land cover map for Germany for several years
  • Calculation of differences between different years
  • Provision of indicators for CODE-DE services
Customer
Federal Ministry of Transport and Digital Infrastructure (BMVI) – mFund

Story

The economical use of land as a resource has been the subject of planning interventions and measures as part of the German government’s sustainability goals for many years. Particularly in dealing with settlement and traffic areas, the demands for use are conflict-ridden. In this problematic situation, the data basis of the monitoring of settlement and transport area development is of particular importance.
The data basis of the Copernicus Sentinel satellites allows to supplement and improve indicators of settlement and transport area development. The quality and informative value of these indicators were researched in the project Inwertsetzung von Copernicus-Daten f├╝r die Raumbeobachtung (incora, BMVI mFund F├Ârderkennzeichen: 19F2079C) and an automatic procedure was developed to generate annual land cover classifications for all of Germany.

terrestris