Files
Abstract
Stress and environmental conditions in broiler production significantly impact animal welfare, growth rates, and disease susceptibility. While traditional methods rely on periodic manual checks of the environment parameters, they often fail to detect early signs of distress. This study introduces a real-time monitoring system that combines IoT sensors and machine learning to enable real-time assessment of broiler welfare. The system integrates environment monitoring data from BME680 sensor, with the Sensiron FSM3200 for air velocity, and with FLIR camera for thermal and RGB imaging. Our machine-learning model leverages keypoint detection and segmentation to precisely analyze body temperatures and continuously track bird health. The collected environmental data is displayed through a web-based dashboard, facilitating real-time insights for temperature regulation and automated interventions. This integrated system offers early detection of potential health risks, ensuring proactive management and improved bird welfare throughout the poultry house and during transportation.