AI to detect invasions of Asian hornets
Asian hornets are known to have invaded mainland Europe and parts of east Asia.
A new automated AI system could be used to detect the presence of Asian hornets, and provide a warning if an invasion approaches a region.
VespAI, developed by researchers at the University of Essex, is able to attract the hornets to a monitoring station, capture images of the insects with an overhead camera, and alert users.
Asian hornets, which are also known as yellow-legged hornets, have previously invaded mainland Europe, as well as parts of east Asia. There have also been invasions reported in the United States of America, in the states of Georgia and South Carolina.
The researchers say that, with the United Kingdom at the edge of the European invasion front, and seeing yearly invasions, the need for an improved monitoring system has become urgent.
They say that the development of VespAI will provide a robust and accurate warning system to detect new ingressions of Asian hornets.
The system operates using a compact processor, and will remain dormant until it recognises an insect within the size range of the hornet. It will then activate its AI algorithm, which analyses an image of the insect to detect whether it is an Asian hornet (Vespa velutina) or a native European hornet (Vespa crabro).
If the insect is identified as an Asian hornet, VespAI will send an image alert to the user and ask them to confirm the identification.
The research team believe that the system will prove resourceful to environment agencies monitoring for invasions. Agencies currently rely on submitted reports, meaning that they have to manually validate thousands of images per year – many of them of misidentified species.
Other parts of Europe use trapping to detect hornets, which can result in many native insects dying while having little impact on Asian hornet levels.
During testing on the island of Jersey, which sees high numbers of Asian hornet incursions, VespAI proved accurate in identifying Asian hornets. While exposed to both Asian hornets and European hornets, as well as a variety of other insects, the system was able to distinguish between each of the insects.
Dr Peter Kennedy, who conceptualised VespAI, said: “VespAI does not kill non-target insects, and thus eliminates the environmental impact of trapping, while ensuring that live hornets can be caught and tracked back to the nest, which is the only effective way to destroy them.”
Dr Thomas O’Shea-Wheller, from University of Exeter’s Environment and Sustainability Institute, said: “That’s the benefit of our system – its high accuracy means that it won’t wrongly identify other species, or miss any Asian hornets that visit,”
The full study can be found in the journal Communications Biology.
Image © Shutterstock
