by The island of Jersey is famous for its picturesque scenery and mild climate. However, its proximity to continental Europe has placed it in the path of an invasive predator: the Asian hornet, Vespa velutina. This is bad news for the island’s wildlife, but Jersey is a great location to test new methods of dealing with the horns.
Using the island as a test bed, our team of biologists and data scientists have developed an AI system that can automatically detect invasive Asian hornets in new regions, enabling them to be eradicated before they can gain a foothold. . The system, VespAI, identifies Asian hornets with more than 99% accuracy, as shown by recent results published in the journal Communications Biology.
This early warning system is necessary because the Asian horn – also known as the yellow horn – is a rapidly spreading threat to biodiversity. Originally from southeast Asia, the species has invaded Europe, East Asia and the United States within two decades, preying on honey bees and other important pollinators.
To tackle the issue, VespAI uses deep learning technology, a computational method inspired by the human brain, to visually identify Asian hornets, capture images and send automated alerts to the beekeeper, landowner or government official used by the system.
The setup is simple. A sponge filled with hornet attractant – a delicious mixture of sugar and fruit extracts – is placed under a small camera that takes pictures of visiting insects as they land for food. The system’s AI algorithm then analyzes these images, allowing the monitor to identify Asian horns with near-perfect accuracy.
VespAI fulfills an urgent need for the UK, as Britain sits on the brink of Europe’s invasion of the horns. Asian hornet invasions into the UK began in 2016 and now occur on an annual basis. Although the national eradication strategy has restricted the ability of hornets to establish completely, 2023 was a particularly bad year, with a record 72 nests found.
In Europe, traditional methods such as baited traps are the first line of defence. But these kill significant numbers of native insects including many pollinators. Furthermore, kill traps do little to significantly reduce the number of wasps, as the colonies continue to survive unless the queen is destroyed.
By avoiding harm to visiting insects, VespAI offers a critical advantage over traps. Also, any Asian horns that the system detects remain alive, so they can be traced back to the nest. This is important, as it is the only proven way to destroy colonies.
A key reason for the success of branches is the difficulty of detecting their initial invasion of new regions. New colonies have only one queen, but can produce hundreds of additional queens within a year, causing numbers to grow exponentially.
As a result of this rapid expansion, horned populations are often established before or soon after they are noticed by humans, making eradication almost impossible. As early detection of an invasion depends on seeing the public, there is also a high risk of mistaken identity. In the UK, only 0.06% of the thousands of Asian hornet reports submitted annually are correctly identified.
A scalable solution
To test the effectiveness of VespAI, we conducted rigorous field trials on the island of Jersey, where Asian hornet invasions from France are common. Working with the Jersey Asian Hornet Group and the island government, we implemented prototype systems in areas where Asian hornets, European hornets and other insects were abundant.
We then carefully analyzed the thousands of images collected by the prototype systems. We checked each one manually to make sure the AI algorithm correctly identified all Asian hornets without misidentifying other insects as hornets.
So far, our results are promising. VespAI correctly detects Asian hornets with an accuracy of more than 99%, which provides a significant advantage over previous methods. Although this technology is still in the development stage, our experiments show that the system could transform the way Asian hornets are managed in endangered regions such as the UK and USA.
In particular, the first confirmed Asian hornet of 2024 was recorded in Britain in March, much earlier than usual, and the high number of nests in 2023 suggests that this year may be the most difficult to until now. Further trials of the VespAI system will take place in the UK over the coming months, as we work with government authorities to support eradication efforts at this critical juncture.
This article from The Conversation is republished under a Creative Commons license. Read the original article.
Thomas O’Shea-Wheller receives funding from UK Research & Innovation (UKRI).
Peter Kennedy receives funding from UK Research & Innovation (UKRI) and the Halpin Trust.
