In the heart of Tennessee, a team of researchers led by Yanqiu Yang from the University of Tennessee’s Department of Animal Science has developed a cutting-edge tool that could revolutionize the way we combat invasive species, with significant implications for the energy sector. The tool, named bioWatch, is an AI-powered mobile application designed to detect and monitor the Spotted Lanternfly (SLF), an invasive species that poses a substantial threat to ecosystems and economies worldwide.
The Spotted Lanternfly, native to Asia, has been wreaking havoc on crops, forests, and vineyards since its arrival in the United States. The insect feeds on the sap of plants, weakening and eventually killing them. In the energy sector, the SLF can cause significant damage to power lines and other infrastructure, leading to costly repairs and potential power outages. “The Spotted Lanternfly is a serious threat to our infrastructure and economy,” says Yang. “Our goal with bioWatch is to provide a tool that can help us stay ahead of this invasive species and mitigate its impact.”
bioWatch leverages the power of artificial intelligence, specifically a deep learning model called YOLO11n, to classify SLF across three life stages: adult, early nymph, and late nymph. The model was trained on a dataset of 1161 annotated images, achieving impressive results. “The model’s performance was particularly robust for adults and late nymphs, with recalls of 0.730 and 0.896, respectively,” explains Yang. “This means that the model is highly effective at identifying these life stages, which is crucial for early detection and prevention.”
The application also integrates geospatial mapping, augmented reality, and explainable AI visualization to enhance user engagement and understanding. “We wanted to make the tool as user-friendly and informative as possible,” says Yang. “The geospatial mapping feature allows users to see where SLF have been spotted, while the augmented reality feature provides a visual guide for identifying the insect. The explainable AI visualization helps users understand how the model makes its predictions, which can be particularly useful for educational purposes.”
Early user feedback has been overwhelmingly positive, with many praising the app’s potential for biodiversity monitoring and K–12 education. “The app is not only a powerful tool for invasive species detection, but it also has the potential to inspire the next generation of scientists and conservationists,” says Yang.
The research was recently published in the journal ‘Smart Agricultural Technology’, which translates to ‘Intelligent Agricultural Technology’ in English. The publication highlights the growing importance of technology in agriculture and the need for innovative solutions to combat invasive species.
The implications of this research are far-reaching. As invasive species continue to threaten ecosystems and economies worldwide, tools like bioWatch could play a crucial role in their detection and management. In the energy sector, the app could help prevent costly damage to infrastructure and ensure the reliable delivery of power.
Looking to the future, Yang and her team are already planning to expand the app’s capabilities. “We see bioWatch as a modular platform that can be updated and improved over time,” says Yang. “We’re excited to explore new features and applications that can further enhance its effectiveness and reach.”
In the fight against invasive species, bioWatch represents a significant step forward. By harnessing the power of AI and engaging users in the process, it offers a promising solution to a growing problem. As the energy sector continues to grapple with the impacts of invasive species, tools like bioWatch could prove invaluable in mitigating their effects and ensuring the reliable delivery of power.