The research funding award "Program Grant" of the international "Human Frontier Science Program" (HFSP) in the selection round 2021 goes to Prof Benjamin Risse from the Department of Mathematics and Computer Science at the University of Münster. The computer scientist receives the three-year grant for a project to research new AI (artificial intelligence) algorithms and imaging systems to study plant-pollinator interactions. He is implementing the project together with an international team.
The selection process for HFSP funding is highly competitive. Out of an original 709 applications from researchers from more than 50 countries, 28 projects are now funded after a multi-stage selection process - seven in the "Early Career" funding line and 21 in the "Programme Grants" line. Each team member will receive between 110,000 and 125,0000 US dollars per year for three years, which is roughly equivalent to 300,000 euros in total.
An international team consisting of plant geneticists (Sweden), insect researchers (USA) and computer scientists from Benjamin Risse's group will investigate plant-pollinator interactions. These interactions usually serve both plants and animals by enabling pollination and nectar uptake and thus play an important role in the functioning of our ecosystems. Among other things, the scientists would like to use this approach to gain insights into more sustainable and efficient agricultural food production methods. "Especially in times of global warming and insect defaunation, these interactions will play an important role in the future," says Benjamin Risse. Through a close interdisciplinary collaboration, the scientists take into account different factors of plant-pollinator interactions to thereby develop efficient approaches to these challenges. In particular, Benjamin Risse's working group will be responsible for developing the recording and analysis tools that will enable colleagues from biology and ecology to quantitatively study the insects' behaviour. "The computer-based analysis of small, often indistinguishable objects such as insects in dynamic environments is not readily possible to date and poses a great challenge to existing state-of-the-art AI algorithms," explains Benjamin Risse. The research of the computer scientists from Münster therefore focuses on the limits of current AI methodologies for image analysis. The scientists hope that the methods developed in the project can be used for many other areas of application in the field of sustainability research in the future.