The Swiss Federal Institute of Technology in Zurich (ETH) and Swiss Federal Institute of Technology in Lausanne (EPFL) have developed the AI-based software Pomelo for humanitarian operations. To this end, they teamed up with the International Committee of the Red Cross (ICRC) und the Qatar-based Hamad Bin Khalifa University. According to a press release issued by EPFL, this software is the first to provide digital maps for population density per hectare with the required level of accuracy. This information is important for operational planning, for example when it comes to restoring the water supply, distributing food or implementing a prevention program. The team’s findings were recently published in the journal Scientific Reports.
“There are a few other artificial-intelligence-based programs out there, but they all need a precise census count to start learning, which they then refine with other data”, explains Devis Tuia, who heads up the Environmental Computational Science and Earth Observation Laboratory at EPFL, in the press release. However, such data is rarely available in locations where the ICRC operates and, where it does exist, is often outdated. “We only need an estimate of the population at the coarse regional level”, Tuia adds.
On this basis, Pomelo brings together large amounts of publicly available data such as building numbers, average building sizes, proximity to roads, road maps and information on night lighting. After this, the software aggregates this data on the basis of weightings learned from a neural network. According to the information, Pomelo has already been successfully tested in countries such as Tanzania, Zambia and Mozambique.
Related news
Meet with an expansion expert
Our services are free of charge and include:
- Introduction to key contacts in industry, academia, and government
- Advice on regulatory framework, taxes, labor, market, and setting up a company
- Custom-made fact-finding visits, including office and co-working space