Using AI to detect Asian hornets
Researchers at the University of Exeter, England, have developed a device that uses artificial intelligence (AI) to identify Asian hornets. AI is when a computer is able to learn from experience and quickly perform tasks that would normally take humans a very long time.
Asian hornets are an invasive species, meaning they are not native to the UK and can cause problems in our ecosystem (the environment and its community of living things). They eat native insects and are especially a threat to honey bees. Just one Asian hornet can kill up to
50 honey bees in a single day.
The new system, called VespAI, uses a sponge cloth soaked with a liquid designed to attract hornets. Once the hornets have arrived, an automatic camera takes pictures of them. AI works out if the insect in the picture is an Asian hornet or a European hornet, a native species.
If an Asian hornet is detected, an alert is sent to a researcher’s phone or computer. VespAI acts as an early-warning system to detect if Asian hornets have made their way into a new region of the UK. It is very accurate and allows the authorities to spring into action quickly to track the Asian hornets, find their nest and destroy it before more are born.
Experts are becoming increasingly concerned that Asian hornets have made a permanent home in the UK – rather than just flying here or being blown across the sea. At the moment the UK’s method for dealing with Asian hornets involves people seeing, identifying and reporting them. However, most of these reports turn out to be other species, so it isn’t a very good way of doing things.
Dr Thomas O’Shea-Wheller, who worked on VespAI, said this new system won’t wrongly identify other species or miss Asian hornets that visit. VespAI has been tested on the island of Jersey and will be tried out in other locations this year.