An Automated Method for Monitoring Footpad Health of Cage-Free Hens
ISSN 2158-9429
Volume 18, Issue 1 - June 2025
Editor: Linda Chalker-Scott
Abstract
Footpad lesion or dermatitis is a common poultry condition that can negatively influence chickens’ production, welfare, and health. However, no automated tool for monitoring FPD in live chickens is currently available. Besides broiler chickens, footpad dermatitis could happen in broiler breeders and cage-free egg layers as well. The objective of this study was to develop and optimize deep learning models to monitor hens’ footpad health including footpad dermatitis and bumblefoot. In recent years, the YOLO (You Only Look Once) family models have gained significant prominence due to their exceptional speed and accuracy in object detection tasks. In this study, the YOLOv5x, YOLOv5s, and YOLOv5m were trained for bumble foot detection (BFD) and YOLOv7 and YOLOv8 models were extracted from GitHub Ultralytics for analyzing FPD. The new models were tested in cage-free layer facilities. By using deep learning models, the precision of bumblefoot detection and the footpad dermatitis detection reached 93.7% and 95%, respectively. The results show that the YOLOv8l outperformed other models in FPD detection, with higher recall (96.6%), mAP@0.50 (97.0%), and F1-score (95.0%).
View/Download Full Article