Back

Fast machine learning with EO data with new Sentinel-2 collections

 

Have you ever wanted to perform a fast prototyping of your machine learning models on a global scale?

Now you can even produce a Sentinel-2 time-lapse of the entire world for a full year  in a fast cost-effective manner thanks to a newly created 120-meter resolution harmonized yearly stack that can be used by everyone.
 
 
The new collections, also available on CREODIAS, can facilitate the creation of various Machine Learning workflows with EO imagery from Sentinel-2 on a global scale. This can be applied in many areas of research or education, e.g. in agriculture (crop classification), climate change (snow cover) and many more. Now you can perform these analyses yourself, in EO Browser, Jupyter Notebook or elsewhere.
 
Stack details
 

Source

Sentinel-2 L2A

Temporal Availability

10-daily periods in 2019

Geographical Area

Land surface area between 58 degrees South and 72 degrees North

Resolution

120 m

Band Information

B02 (blue), B03 (green), B04 (red), B08 (NIR), B11 (SWIR), B12 (SWIR)

Format

COGs tiled by UTM Zones

 

New collection can be used in various machine learning workflows, to run them on a global scale like land cover, crop classification, yield prediction and many others.

This activity was performed in the scope of Horizon 2020 Global Earth Monitor project and the European Space Agency’s Digital Twin activities.

Credit: Sinergise/Sentinel Hub

A description of the data and how to use it on the CREODIAS platform is available here.

Full dataset is available in CREODIAS Platform directly from cloud resources in the EODATA catalogue (over s3):

bucket: DIAS
prefix: auxdata/S2_L3_MOSAIC_120/
 
More information from the authors of this dataset can be found at this link: