Real-time Deer Detection and Warning in Connected Vehicles  

Deer-vehicle collisions represent a critical safety challenge in the United States.  This tool is a real-time detection and driver warning system that integrates thermal imaging, deep learning, and vehicle-to-everything communication to help mitigate deer-vehicle collisions. Our system was trained and validated on a custom dataset of over 12,000 thermal deer images collected in Mars Hill, North Carolina.  Experimental evaluation demonstrates exceptional performance with 98.84% mean average precision, 95.44% precision, and 95.96% recall. The system was field tested during a follow-up visit to Mars Hill and readily sensed deer providing the driver with advanced warning. Field testing validates robust operation across diverse weather conditions, with thermal imaging maintaining 88-92% detection accuracy in challenging scenarios where conventional visible-light based cameras achieve less than 60% effectiveness.  When a high probability threshold is reached sensor data sharing messages are broadcast to surrounding vehicles and/or roadside units via cellular vehicle to everything (C-V2X) communication devices. Overall, our system achieves end-to-end latency consistently under 100ms from detection to driver alert. This research establishes a viable technological pathway for reducing deer-vehicle collisions through thermal imaging and connected vehicles. While the system evaluation provides proof of concept, a primary limitation of the system is the detection range due to resolution.  Further testing with higher resolution cameras will improve the system’s utility.

Link to the tool: TBD

Related Publication: H. Puppula, W. Sarasua,  S. Biyaguda, F. Farzinpour, M. Chowdhury.” Real-time Deer Detection and Warning in Connected Vehicles via Thermal Sensing and Deep Learning” 2026 TRB Annual Meeting, under review.