Penn State Tests Drone Technology for Turkey Welfare

By Blake Jackson

Turkeys take center stage in households across the country, researchers at Penn State have made progress on a technology that could transform how poultry farmers monitor their flocks. A team led by Penn State animal scientist Enrico Casella has successfully tested a new method that uses drone technology and artificial intelligence to observe and record turkey behavior more efficiently on large farms.

Monitoring poultry health and activity is vital for both productivity and animal welfare, but traditional methods require significant labor and time. To address this, the researchers explored whether a lightweight drone equipped with a camera and computer vision software could detect and classify turkey behaviors automatically.

According to Casella, assistant professor of data science for animal systems in the College of Agricultural Sciences and member of the Penn State Institute of Computational and Data Sciences, this is the first known study to evaluate drone-assisted behavior tracking for commercial poultry.

“This work provides proof of concept that drones plus AI can potentially become an effective, low-labor method for monitoring turkey welfare in commercial production,” Casella said. “It lays the groundwork for more advanced, scalable systems in the future.”

The research team recorded footage four times daily using a standard camera drone as it flew over 160 young turkeys ranging from five to 32 days old at the Penn State Poultry Education and Research Center.

Individual video frames were extracted and labeled to build a dataset of more than 19,000 images representing behaviors such as feeding, drinking, sitting, standing, huddling, wing flapping, and perching.

The dataset was used to train and test a YOLO (You Only Look Once) computer vision model, a tool commonly used for object recognition tasks. After evaluating several versions, the researchers found the best-performing model identified 87% of behaviors present and classified actions with 98% accuracy impressive results considering the complexity of real barn environments.

“The study shows that a drone equipped AI system can accurately detect turkey behaviors,” Casella explained. “This method could reduce labor demands, it could allow continuous, non-invasive monitoring of bird welfare in commercial farms and it may also reduce the need for constant human presence, lowering training and staffing burdens.”

Photo Credit: istock-peopleImages