Read the BETTER report about the event here!
After the successful Joint EO Hackathon spearheaded by BETTER last year in Frascati, we are now finalizing a second BETTER standalone event. Given the circumstances and travel restrictions, we will offer an interactive workshop in place of a physical hackathon in the form of a series of independent webinars, supported by technical BETTER consortium members. The advantage of this setup is that you can participate from the comfort of your home/office, at zero cost!
The individual exercises are completely independent and do not build on each other – such that participants are free to register and participate only in the ones that are of individual interest. The exercises will be discussed on the platform GoToWebinar. The step-by-step exercises will feature the use of Binder and its purposely provided cloud resources.
The exercises, for which preparatory material and requirements will be circulated in advance (including, where necessary a pre-recorded preparatory webinar), can be carried out as a team or individually. After an appropriate time for participants to carry out the tasks, solutions will be discussed jointly.
1. SAR-based change detection: enhanced analysis and postprocessing
Exercise owner: Consortium member SatCen
Description: The goal is to identify classification from a SAR RGB. The complexity of the exercise will depend on the audience skills (from a simple threshold application to a clustering algorithm).
Requirements: This step-by-step exercise, supported by Terradue, will feature the use of Binder and purposely provided cloud resources.
Target Participants: Students, Software developers, Data scientists, EO-developers, anyone with an interest in the topic.
Time requirements: Half day (4 hours)
Date & Time: 22 October 9:00am – 1:00pm CEST
Register for this exercise Here!
2. Comparing static and dynamic effects of earthquakes on the ground surfaces
Exercise owner: Consortium member ETHZ
Description: The focus of the exercise is to find a convenient way of exploitation of Co-seismic interferograms. The idea is to produce geocoded maps combining important results.
Requirements: This open exercise, supported by Deimos and Terradue, will feature the use of Binder and purposely provided cloud resources.
Target Participants: Students in remote sensing / geophysics/ data sciences.
Time requirements: Half day (4 hours)
Date & Time: 22 October 1:30pm – 5:30pm CEST
Register for this exercise Here!
3. Enhancing Agricultural Mapping with BETTER Pipelines
Exercise owner: Consortium member WFP
Description: The exercise will evaluate the impact of the usage SAR and of filtered S2 data upon the quality of crop type mapping outputs against a benchmark based on unfiltered S2 data only.
Requirements: This open exercise, supported by Terradue, will feature the use of Binder and purposely provided cloud resources.
Target Participants: Students, Software developers, Data scientists, EO-developers, anyone with an interest in the topic.
Time requirements: Whole day (8 hours)
Date & Time: 23 October 9:00am – 5:00pm CEST
Register for this exercise Here!
4. Semantic Geo-Clustering with SANSA
Exercise owner: Consortium member Fraunhofer IAIS
Description: The use of required semantic transformation and SANSA (Scalable Semantic Analytics Stack - http://sansa-stack.net/) libraries will be carried out on a sample.
Requirements: This step-by-step exercise, supported by external collaborator MKLab ITI-CERTH (EOPEN project) requires the setup of local Docker images.
Target Participants: Students, Software developers, Data scientists, EO-developers, anyone with an interest in EO and semantic interoperability topics;
Time requirements: Half day (4 hours)
Date & Time: 29 October 1:30pm – 5:30pm CET
Register for this exercise Here!
For more details stay tuned via our social media and website - we are looking forward to your participation!