Drones, AI and ducks: How Mizzou is leading the future of wildlife conservation

[Link]These boxes are part of the advanced image analysis after the drone images are collected. Green boxes represent true bird predictions, yellow boxes represent false positive predictions and red boxes represent false negative predictions. Oct. 22, 2025Contact: Eric Stann, [email protected] courtesy Yi Shang Above Missouri's wetlands, a drone quietly hums above flocks of migratory water birds, its camera capturing the ripples of movement below. With this technology, University of Missouri researchers are redefining how wildlife is studied and protected. For decades, scientists have relied on airplanes to count birds - a method that's loud, costly and sometimes dangerous. Accuracy depends on the human eye, and even trained observers can sometimes miss details when birds scatter or blend into their surroundings. Now, scientists at Mizzou's College of Engineering, led by Professor Yi Shang, are taking that process to new heights. By pairing drones with artificial intelligence, the team developed a smarter, safer and faster way to track the migration patterns of these birds - including mallards and pintails, two species of wild duck common to Missouri. Their efforts could transform how this conservation work is carried out across the state and beyond. How it works To put their approach into practice, the Mizzou team uses a combination of flight planning and advanced image analysis. Using specialized software, researchers plot the drone's flight path, identifying the best settings for altitude, speed and image overlap. AI algorithms then analyze the photos, identifying individual birds and preventing double counts. The software can tell whether the birds are on the water, in vegetation or in fields - and can even identify different species, giving scientists a clearer view of the ecosystem. "For straightforward situations, such as birds on open water, the technology is more than 95% accurate," Shang, professor of electrical engineering and computer science and Robert H. Buescher Faculty Fellow, said. "Even in complex circumstances - where birds either overlap or are covered by trees or crops - our method is still 80-85% accurate." The system then pairs with large language models to analyze the images and generate easy-to-read summaries, giving the Missouri Department of Conservation and other agencies useful insights into the movements of migrating birds. In the future, drones are expected to become more affordable and come equipped with higher-resolution cameras, allowing for better detection and classification methods. And the technology isn't limited to counting birds - it could monitor other species and ecosystems, similar to how AI is already being used to analyze images of deer and other wildlife from game cameras. With their innovative approach, Mizzou researchers are helping ensure that wildlife conservation soars to new heights - and that migratory birds across Missouri are tracked safely, accurately and efficiently. The study, "New methods for waterfowl and habitat survey using AI and drone imagery," was published in the journal Drones.

MU College of Engineering

Facebook Share Linkedin Post