AI Cameras Monitor Mithun Herds in Nagaland

Kohima: Researchers at the ICAR-National Research Centre on Mithun have deployed an artificial intelligence system to track the behavior of Mithun, also known as the Cattle of the Hills. The project uses 12 high-definition CCTV cameras across two sheds at the farm. These cameras provide infrared coverage for 24-hour monitoring. The system eliminates the need for manual, round-the-clock observation.

The tech team trained the software using 3,000 annotated images. These images show four specific behaviors: feeding, standing, lying, and mounting. By combining the YOLOv8n object detection model with DeepSORT tracking, the system assigns unique identities to individual animals. It processes 31 frames per second on an NVIDIA RTX 3060 GPU. The model hit 99.5% mean average precision and 99.6% recall during testing.

It worked despite shadows, motion blur, and uneven ground. The study, published in Engineering Research Express, was a collaboration between ICAR-NRC on Mithun, NIT Nagaland, Nagaland University, and CHRIST (Deemed to be University). Current human observation methods often fail at night. An automated system could provide farmers with vital data on health and reproductive needs.

The team knows there is more work ahead. “The technology is still at an early stage,” the researchers noted. Future versions might track aggression, grooming, or sickness. They also plan to test the gear across different seasons and farms to ensure it stays accurate in varied settings.