Smart neck collar: IoT-based disease detection and health monitoring for dairy cows
Devinder Kaur, Amandeep Kaur Virk
Abstract
The dairy industry is crucial for meeting global milk demands, making the health and well-being of dairy cows paramount. This research paper introduces an innovative approach to monitor the vital health parameters and overall health status of dairy cows through the development of an Internet of Things (IoT)-based neck collar device. This smart collar system incorporates a range of sensors for measuring temperature, pulse rate, and activity level. These sensors continuously collect data, which is wirelessly transmitted to a central database for real-time monitoring. Machine learning algorithms analyse these data to detect anomalies and patterns indicative of various health conditions. By enabling early disease detection, the IoT-based collar facilitates timely intervention, reducing the risk of cow morbidity and mortality, ultimately leading to enhanced milk production and overall herd health. In this paper, the design and architecture of the neck collar has been presented. The smart neck collar is characterized by low power consumption, miniaturization, intelligence, ease of operation, cost-effective new materials, portability, and high performance. By providing continuous and comprehensive monitoring, this technology significantly contributes to dairy cow welfare and farm operational efficiency. A comprehensive dataset containing vital health parameters of 150 dairy cows was collected over a period of two-months from seven districts in Punjab, India, including Fatehgarh Sahib, Ludhiana, Nawan Shahr, Patiala, Sangrur, Ropar, and Hoshiarpur, using smart neck collars. This dataset was analyzed using various machine learning algorithms, including Random Forest, Support Vector Machines, K-Nearest Neighbor, Naïve Bayes, and Decision Trees. The analysis revealed that Random Forest and Decision Trees were the most effective algorithms for predicting the health status of cows, achieving accuracies of 92% and 91%, respectively.