Field Delineation

derived data field delineation on-demand sentinel hub sentinel-2


An algorithm for automatic delineation of agricultural field boundaries from Sentinel-2 imagery.The main part of the algorithm is an advanced pre-trained machine learning model. It was trained on multiple locations throughout Europe for a time interval from March to August.The algorithm is an improved version of the one developed for NIVA project. More information about the process is available in a blog post and in a recording of a webinar.To run the field delineation process, navigate to EDC Browser, select the required input parameters and follow the check-out wizard to complete your order. Produced data will be uploaded directly into your JupyterLab environment on EDC.

Additional info



Geographical coverage

Global coverage but intended for areas with agricultural fields

Temporal availability

January 2016 - ongoing

Update Frequency


Attributes Information

aoiArea of interestPolyon or bounding box
time_intervalA time range of Sentinel-2 data to processString
maxccMaximum cloud coverage of Sentinel-2 data.Float from interval `[0, 1]`

Produced results

results.gpkgA Geopackage containing delineated polygons of fields from selected AOI. Polygons are projected to WGS84 coordinate reference system.


Prices are defined based on amount of data required to process. Check here for more info about pricing and restrictions.


Creative Commons Attribution 4.0 International License. Attribution: Contains modified Copernicus Sentinel data (year) processed by Sentinel Hub


Sentinel Hub

Managed By

Sentinel Hub

See all datasets managed by Sentinel Hub.


Edit this collection entry on GitHub