A Monitoring System for Cattle Behavior Detection using YOLO-v8 in IoT Environments

The decline and aging of the rural population have led to the development of smart livestock farming systems that can automatically detect animal behavior. Previous studies have primarily focused on detecting behavior during the day, cattle mounting behavior can occur at any time, both day and night. This study proposes a continuous 24-hour real-time monitoring IoT system for detecting cattle mounting behavior. This system leverages the convolutional neural network-based YOLO-v8 model to discern between typical cattle behavior and mounting actions. The experimental results demonstrate the robust performance of the integrated day and night model, ensuring accurate predictions.