GMES & Africa Workshop for Services Design | Cloud Computing Training

During this closed session, organised as part of the Joint Research Center (JRC) activities, participants will become familiar with cloud computing and gain hands-on experience on the use of RUS Copernicus virtual machines for Earth Observation projects exploiting Sentinel-1 and Sentinel-2 data for agricultural applications.

The exercises will be based on Open tools (ESA SNAP Toolbox, QGIS, R and Orfeo Toolbox) and will be demonstrated by a trainer connected via internet to a remote RUS virtual machine.


Programme

21 February Morning:
Introduction to basic concepts of cloud computing and its benefits. New trends and challenges in EO and how cloud computing can benefit the full exploitation of data. Examples of cloud computing platforms related to ESA (RUS, DIAS, sen2Agri, TEPs, etc.) will be shown.

21 February Afternoon:
NDVI monitoring of agricultural fields (R – Orfeo/QGIS) Exercise 1
The exercise will demonstrate the methodology to derive a temporal profile of NDVI over agricultural areas. For this, a Sentinel-2 image will be segmented to extract field boundaries. This layer will be used to derive the mean NDVI value per field and its temporal evolution will be analysed.

A large dataset of Sentinel-2 images will be used, highlighting the advantages of using cloud computing for projects dealing with large datasets. Participants will derive a final and lighter product that summarizes all the spectral information retrieved with the time-series analysis.

Participants will run the exercise together with the trainer.

22 February Morning:
Crop mapping with Sentinel-2 (SNAP) Exercise 2
The exercise will demonstrate the methodology to run a supervised classification (Random Forest) to identify different crops in a study area located in the south of Spain (Seville). Sentinel-2 level 2A data will be used and the processing will include a multi-temporal approach.

The methodology will be first demonstrated by the trainers. During the rest of the session, participants will be asked to repeat the exercise by themselves using a step-by-step guide and the assistance of the trainers.

22 February Afternoon:
Joint use of Sentinel-1 and Sentinel-2 for agricultural mapping (R) Exercise 3
The exercise will demonstrate the methodology to run a supervised classification (Random Forest / SVM) to identify different crops in a study area located in the south of Spain (Seville). Sentinel-1 GRD products will be combined with Sentinel-2 level 2A data to demonstrate how the joint use of optical and SAR data can benefit. For the exercise, the GPT version of SNAP and an R script will be used.

A large dataset of Sentinel-1 GRD and Sentinel-2 level 2A images will be used, highlighting the advantages of using cloud computing for projects dealing with large datasets and with high processing requirements. Demonstration of how IT resources can be quickly adapted to the project needs.

Participants will run the exercise together with the trainer.

Training date

Thursday 21st of February 2019 - Friday 22nd of February 2019

Application opening - closing

-
Application is closed

Do you want to receive RUS news and new training info?

Subscribe to our mailing list