AI Revolutionizes Livestock Management with Real-Time Mithun Tracking
In a groundbreaking advancement for livestock management, researchers at the ICAR-National Research Centre on Mithun (ICAR-NRC on Mithun) in Nagaland have developed a sophisticated artificial intelligence (AI) system designed to detect and track Mithun behavior in real time. This innovative framework promises to enhance the monitoring of animal health and welfare, thereby contributing to effective breeding and reproductive management.
The research, recently published in Engineering Research Express, introduces a non-contact approach to continuously track Mithun (Bos frontalis) behavior in natural farm settings. Known as the ‘Cattle of the Hills,’ Mithun plays a vital role in the livelihoods and food security of tribal communities in Northeast India, making reliable behavior monitoring critical to sustaining these practices.
Innovative Monitoring Solutions
Understanding livestock behavior is essential for effective animal management. Variations in behaviors such as feeding and resting can be early indicators of an animal’s health and wellbeing. Traditionally, these observations have relied on labor-intensive manual monitoring, which poses challenges, especially during nighttime hours. The new AI system effectively addresses these limitations.
The research team utilized 12 high-definition CCTV cameras placed in two sheds at the ICAR-NRC on Mithun farm. These cameras provided comprehensive coverage, including infrared capabilities for nocturnal monitoring. The team annotated a dataset of 3,000 images showcasing four essential Mithun behaviors: feeding, standing, lying, and mounting.
Advanced AI Performance
Employing the YOLOv8n model for behavior detection alongside DeepSORT technology, the system can track individual animals and assign them unique identifiers across video frames. This level of innovation led to the model achieving a remarkable mean average precision of 99.5% and a recall rate of 99.6%. The processing speed was approximately 31 frames per second, demonstrating its potential for real-time applications even under challenging conditions like background clutter and nighttime lighting.
Continuous monitoring through this system could transform livestock farming by providing timely behavioral insights that inform animal care strategies, making it easier to recognize issues without requiring constant physical observation.
Future Prospects and Innovations
While the system has undergone initial evaluation at a single farm, researchers emphasize the need for further validation in varied environments and conditions. Current capabilities focus on four behaviors, and the team aims to enhance the framework to include other critical behaviors such as aggression and grooming. Plans for future research involve using temporal AI models and expanding datasets to incorporate multiple farms and seasonal variations.
This study illustrates the exciting potential of merging artificial intelligence with livestock science, paving the way for data-driven approaches to cattle management. By adopting such technology, farmers can benefit from precise, continuous monitoring that significantly improves animal wellbeing and farm productivity.
The full study is available in Engineering Research Express, Volume 8 (2026), Article 175213. It was conducted in collaboration with researchers from NIT Nagaland, Nagaland University, and CHRIST (Deemed to be University).
Observer Voice is the one stop site for National, International news, Sports, Editor’s Choice, Art/culture contents, Quotes and much more. We also cover historical contents. Historical contents includes World History, Indian History, and what happened today. The website also covers Entertainment across the India and World.
Follow Us on Twitter, Instagram, Facebook, & LinkedIn
