AI keeps an eye on the ‘Cattle of the Hills’ — and it could change how farmers watch their herds
AI-powered cameras at an ICAR farm in Nagaland are learning to track Mithun behaviour, potentially giving farmers a new tool for animal health, welfare and reproductive management.
Researchers at the ICAR-National Research Centre on Mithun (ICAR-NRC on Mithun), Nagaland, have developed an AI-based, non-contact system that can automatically detect and track individual Mithun and identify their behaviour in a natural farm environment.
NEW DELHI: Artificial intelligence is being used to monitor the behavior of Mithun, the distinctive livestock species of Northeast India, in real time at an ICAR research farm in Nagaland, in a development that could eventually help farmers improve animal health, welfare and reproductive management.
Researchers at the ICAR-National Research Centre on Mithun (ICAR-NRC on Mithun), Nagaland, have developed an AI-based, non-contact system that can automatically detect and track individual Mithun and identify their behavior in a natural farm environment.
Mithun, also known as the “Cattle of the Hills”, has significant social, cultural and economic importance among tribal communities across the Northeast and contributes to livelihoods and food security.
Cameras meet artificial intelligence
The researchers installed 12 high-definition CCTV cameras across two sheds at the ICAR-NRC on Mithun farm. The cameras provide continuous day-and-night surveillance, including infrared coverage.
From the footage, the team created a dataset of 3,000 manually annotated images covering four behaviours — feeding, standing, lying and mounting.
The AI system combines the YOLOv8n model for detecting behavior with DeepSORT technology to track individual animals and assign persistent identities across video frames.
The detection model achieved a mean average precision of 99.5% at mAP@0.5, with a recall of 99.6%. It processed footage at about 31 frames per second using an NVIDIA RTX 3060 GPU, indicating its potential for real-time deployment.
The system was also tested under challenging conditions, including partial occlusion, background clutter, uneven and wet ground, shadows, motion blur and nighttime infrared footage.
From health to breeding
Changes in feeding, standing and lying patterns can provide indications of an animal’s health, comfort, nutrition and physiological condition. Mounting behavior can provide useful information for reproductive and estrus management.
Automated monitoring could therefore reduce dependence on continuous manual observation, particularly during night hours.
However, the researchers caution that the system has so far been evaluated at a single farm. Further validation across farms, seasons, geographical regions, stocking densities and different camera arrangements is required.
The team plans to expand the system to detect behaviors such as aggression, grooming and disease-related inactivity, while exploring temporal AI models, edge-device deployment and larger datasets.
The study was published in Engineering Research Express, Volume 8 (2026), Article 175213, and involved researchers from ICAR-NRC on Mithun, NIT Nagaland, Nagaland University and CHRIST (Deemed to be University).
