Event-Driven Architecture Masterclass: Engineering a Robust, High-performance...
An automated end-to-end framework for CAP monitoring, On-demand access to the ENVISION Data cube
1. This project has received funding from the European Union's Horizon 2020 research and innovation programme
under grant agreement No 869366.
Legal notice: The ENVISION project and its content reflect only the author’s view, therefore the EASME is not responsible for
any use that may be made of the information it contains!
Thanassis Drivas
National Observatory of Athens – Beyond Centre of Excellence
28/02/2023
2. This project has received funding from the European Union's Horizon 2020 research and innovation programme
under grant agreement No 869366.
Data Cube is the backbone for all our services such as crop mapping, grassland
mowing event detection as it offers an efficient and simple way of organizing the EO
data.
However, additional services and apps can be built on top of it as users have the
potential of directly accessing it.
In that direction, we developed an application that unlocks the power of EO Big Data
to the users by allowing them to ask data from the cube via either an interface or an
API.
The requested data can be analyzed by domain experts, data scientists and decision
makers as the response to the users is sent in multiple formats.
As many organizations use their in-house software for analyzing the EO data and
products, there is the need to ingest ENVISION products into them.
Currently, CAPO is currently exploiting by embedding the requested data in their
systems and thus enhance the validation and photo-interpretation processes.
http://185.178.86.82/
3. This project has received funding from the European Union's Horizon 2020 research and innovation programme
under grant agreement No 869366.
Django web framework has been utilized to create a web
application, along with Django Ninja for developing a REST
API.
Data Available for retrieving:
One or more bands of interest from the Sentinel-2 data,
Vegetation Indices and
Sentinel-1 VV and VH Backscatters,
Data are retrieved from the cube, while on-the-fly processes
such as cloud masking take place.
Afterwards, they are visualized using matplotlib and
seaborn.
The retrieved data can be used to identify features such as
vegetation, water bodies, and urban areas or make several
type of analysis such as monitoring vegetation health,
mapping urbanization, and detecting changes in water quality.
4. This project has received funding from the European Union's Horizon 2020 research and innovation programme
under grant agreement No 869366.
5. This project has received funding from the European Union's Horizon 2020 research and innovation programme
under grant agreement No 869366.
Ask for NDVI Time series for a specific parcel flagged as alert
from the crop classification algorithm
Required Parameters
The minimum cloud free percentage of pixels in the parcel.
The Buffer Zone
The starting date of the observations
The ending date of the observations
The results are ingested into the CAPO's internal system and are
considered a tool to be used for the inspection of the specific parcel
and validation of algorithm's accuracy.
6. This project has received funding from the European Union's Horizon 2020 research and innovation programme
under grant agreement No 869366.
Currently, the web application considered to be an added value tool. However, there is always room for
improvements as the following:
Connect to the ENVISION database giving users the potential for choosing certain category of parcels
(e.g. Alarms from crop classification) in the interface.
Add more vegetation indices based on the CAPO's needs
Add std, mean and median to the same graph
Finalize the xarray-based response from the application
7. This project has received funding from the European Union's Horizon 2020 research and innovation programme
under grant agreement No 869366.
Thanassis Drivas
tdrivas@noa.gr