YOLO-Based Model for Automatic Detection of Broiler Pathological Phenomena through Visual and Thermal Images in Intensive Poultry Houses

The increasing broiler demand due to overpopulation and meat imports presents challenges in poultry farming, including management, disease control, chicken observation varying light conditions. To address these issues, the development of AI-based management processes is crucial, especially considering need for detecting pathological phenomena intensive rearing. In this study, a dataset consisting visual thermal images was created capture broilers. contains 10,000 with 50,000 annotations labeled as lethargic chickens, slipped tendons, diseased eyes, stressed (beaks open), pendulous crop, healthy broiler. Three versions YOLO-based algorithm (v8, v7, v5) were assessed, utilizing augmented image datasets various augmentation methods. aim develop thermal- visual-based models broilers complex environments, secondarily, classify under challenging lighting After training on acknowledged phenomena, YOLOv8-based model demonstrated exceptional performance, achieving highest accuracy object detection (mAP50 0.988) classification (F1 score 0.972). This outstanding performance makes it reliable tool both classification, attributed use comprehensive during development, enabling accurate efficient even environmental By employing visual- thermal-based monitoring, farmers can obtain results from viewpoints, ultimately enhancing overall reliability monitoring process.