ICAR researchers develop real-time system to monitor feeding, standing, lying and mounting behaviour for animal health, breeding and welfare

New Delhi, September 8, 2026: Artificial intelligence is being deployed to monitor the behaviour of Mithun, the culturally and economically important “Cattle of the Hills” found across Northeast India. Researchers at the ICAR-National Research Centre on Mithun (ICAR-NRC on Mithun), Nagaland, have developed an AI-based system that can detect and track Mithun behaviour in real time, potentially reducing the need for continuous manual monitoring.

The technology combines CCTV cameras, computer vision and artificial intelligence to provide day-and-night, non-contact monitoring of animals. The research could support livestock managers and farmers in areas such as animal health, welfare, nutrition, breeding and reproductive management.

Key Highlights

• 12 high-definition CCTV cameras deployed across two sheds at the ICAR-NRC on Mithun farm.
• Cameras provide continuous day-and-night monitoring, including infrared surveillance.
• AI dataset developed using 3,000 manually annotated images.
• System currently detects four behaviours: feeding, standing, lying and mounting.
• YOLOv8n used for behaviour detection and DeepSORT for individual animal tracking.
• YOLOv8n achieved 99.5% mAP@0.5 and 99.6% recall.
• Processing speed reached approximately 31 frames per second using an NVIDIA RTX 3060 GPU.
• System tested under challenging conditions including nighttime infrared footage, motion blur, shadows, wet ground and partial occlusion.
• Future applications could include monitoring aggression, grooming and disease-related inactivity.
• Researchers are exploring edge-device deployment, temporal AI models and larger multi-season datasets.

AI Moves Livestock Monitoring Beyond Manual Observation

Mithun, scientifically known as Bos frontalis, plays an important role in the social, cultural and economic life of tribal communities in Northeast India. Monitoring its behaviour is particularly relevant because changes in feeding, standing and lying patterns can provide indications of an animal’s health, comfort, nutrition and physiological condition.

Mounting behaviour can also provide useful information for oestrus and reproductive management.

Traditionally, such observations depend heavily on human supervision, making continuous monitoring difficult, particularly during night-time. The new AI system seeks to address this limitation by automatically identifying animal behaviour while tracking individual Mithun across video frames.

How the AI System Works

Researchers created a dataset from footage captured by 12 CCTV cameras and manually annotated 3,000 images covering four major behaviours.

The YOLOv8n model performs behaviour detection, while DeepSORT assigns persistent identities to individual animals and tracks them across successive frames. This enables the system to determine not only what an animal is doing but also to follow its behaviour over time.

The system maintained performance even in conditions such as background clutter, uneven or wet ground, shadows, motion blur, partial obstruction and infrared nighttime footage.

From Livestock Monitoring to Precision Farming

The technology could eventually contribute to precision livestock farming, where behavioural data is used to support more informed management decisions.

Automated alerts or behavioural records could help livestock managers identify changes that may otherwise go unnoticed during routine inspections. Over time, larger datasets could allow AI systems to recognise behavioural patterns associated with health problems, reproductive cycles or changes in animal welfare.

However, researchers have noted that the system has so far been evaluated at a single farm. Further validation is required across different farms, regions, seasons, stocking densities and camera configurations. Severe occlusion can also affect detection and tracking performance.

Future research will look at expanding the system to identify aggression, grooming and disease-related inactivity, alongside larger datasets and more advanced AI models.

---

Impact on Hospitality, Food and Travel & Tourism

Although the research is primarily focused on livestock management, it has potential relevance for the wider food, hospitality and tourism ecosystem, particularly in Northeast India.

Food & Livestock Supply Chain

Improved livestock monitoring can support better animal health, welfare and productivity, potentially strengthening the long-term livestock supply chain. For the food sector, technology-enabled management could contribute to more consistent and data-driven livestock production.

Culinary Tourism

Mithun has a distinctive cultural identity in Northeast India. Greater visibility of the animal and its traditional role could contribute to regional food storytelling and culinary tourism, particularly around indigenous cuisines and tribal food traditions.

Rural & Experiential Tourism

Technology such as AI-enabled farm monitoring could also become part of the emerging agri-tourism and rural tourism ecosystem. Demonstration farms and livestock research centres could potentially develop educational experiences around indigenous breeds, traditional farming and modern agricultural technology.

Northeast India Destination Development

Mithun is closely connected with the cultural identity of communities in the Northeast. Combining indigenous livestock, local food traditions, rural livelihoods and agricultural innovation could create differentiated tourism narratives for the region.

Industry Takeaway

The Mithun AI project highlights how artificial intelligence and computer vision are moving into livestock management, bringing continuous behavioural monitoring closer to farmers and research institutions.

For hospitality and tourism, the larger opportunity lies in connecting agriculture, indigenous food, rural entrepreneurship and experiential tourism. As precision livestock technologies mature, they could simultaneously strengthen farm management and create new stories and experiences around India’s distinctive regional food and agricultural heritage.