The Copernicus program is a European Commission and European Space Agency led project for earth observation. The program was officially launched in 2008 and has since then sent various satellites and instruments into space. The main purpose of the satellites is earth observation, i.e., collecting images of our planet. These images capture different frequencies of light and one can compute and classify various emissions of heat, light, and gases.
The Leicester Innovation Hub, together with various international partners, organized a hackathon where participants use Copernicus data to address challenges around food security.
The team that I joined had members from Italy, the UN, a bank, and our University. Two days are fairly short for a new project and some of us spent the first day on just gathering information what is and what is not possible using Copernicus data.
Of course our project would involve some form of machine learning as we wanted to address a specific topic in precision agriculture where off-the-shelf models were not available. However, the use of machine learning usually means that end up with at least two problems instead of one: in addition to finding a model you also need to find the right data to train the model. At the end of day one we had basically decided that our aims were too ambitions and we needed to go for something a lot more basic.
At the end of day two our team delivered their presentation and from initially 8 participating teams, we placed second!
P.S.: The description above is very vague on purpose: some team members have announced interest to develop the project further.
The Leicester Innovation Hub, together with various international partners, organized a hackathon where participants use Copernicus data to address challenges around food security.
Illustration by Airbus, a partner of the Copernicus program
The team that I joined had members from Italy, the UN, a bank, and our University. Two days are fairly short for a new project and some of us spent the first day on just gathering information what is and what is not possible using Copernicus data.
Of course our project would involve some form of machine learning as we wanted to address a specific topic in precision agriculture where off-the-shelf models were not available. However, the use of machine learning usually means that end up with at least two problems instead of one: in addition to finding a model you also need to find the right data to train the model. At the end of day one we had basically decided that our aims were too ambitions and we needed to go for something a lot more basic.
At the end of day two our team delivered their presentation and from initially 8 participating teams, we placed second!
P.S.: The description above is very vague on purpose: some team members have announced interest to develop the project further.
Comments
Post a Comment