Spotting the Japanese beetle with just a few clicks

It is a major threat to hundreds of plant species. Present in Europe since 2014, the Japanese beetle (Popillia japonica) was first spotted at the southern Swiss border in 2017. To prevent it from spreading further in our country and in Europe, Agroscope, the Swiss competence centre for agronomic research, has just launched an innovative interactive online map, developed with the contribution of the Dalle Molle Institute for Artificial Intelligence (IDSIA USI-SUPSI) to record and report the presence of the insect in Ticino.

A threat to agriculture

With its ability to feed on more than 300 plant species, including grasslands and fruit and vegetable crops, the Japanese beetle is a major threat to the agricultural sector. This is well known in the United States, where more than $450 million is spent annually to control it from spreading. Accidentally introduced to Europe in 2014 near Milan, Italy, Popillia japonica can travel easily, thanks also to the movement of goods and people, and can settle quite rapidly in a wide range of different climates, notably across the continental area from the Atlantic to the Black Sea, and from the Mediterranean to Britain and southern Scandinavia. For these reasons, the relevant agencies of the European Commission have named Popillia japonica a high-priority pest in the European Union’s new phytosanitary law. Since 2017, the insect has also been present in Ticino in certain areas on the border with Lombardy.

Artificial intelligence "Made in USI" to support public collaboration

"USI has developed, in collaboration with Agroscope and In-Finitude AG (a spin-off company of ETH Zurich) and as part of a project on, artificial intelligence-based software for the recognition of Popillia japonica, the insect that is particularly harmful to crops and is expanding in Switzerland," explains Prof. Cesare Alippi of IDSIA and full professor at the USI Faculty of Informatics. "The software was then integrated into a collaborative platform through which experts in the field and ordinary citizens can send in reports of the insect’s presence, complete with photographs. The software, which was also developed in collaboration with Dr. Alberto Ferrante , a researcher in my group, is particularly effective in recognising the insect in photographs taken in the natural environment so as to automate its identification and enable models of beetle distribution".

The Pollenn collaborative platform into which the system developed at USI has merged is an innovative tool that allows people to easily report - with just a few clicks - their observations in the field, thus helping to observe the spread of the beetle. To access and contribute: