Real-time tracker of chicken for poultry based on attention mechanism-enhanced YOLO-Chicken algorithm

•A novel chicken monitoring system for poultry industry frees farmers.•Attention mechanism-enhanced YOLO-Chicken algorithm outperforms the others.•BoT-SORT tracking algorithm precisely monitors the track of the chickens.•A graphical interface is developed to visually display tracking status. The mod...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Computers and electronics in agriculture Ročník 237; s. 110640
Hlavní autoři: Jiang, Daqi, Wang, Hong, Li, Tan, Gouda, Mohamed Amin, Zhou, Bin
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier B.V 01.10.2025
Témata:
Attention mechanism
Chicken tracking system
Multi-object tracking algorithm
Multi-scale feature
YOLOv8
Multi-object tracking algorithm
Attention mechanism
YOLOv8
Chicken tracking system
Multi-scale feature
ISSN:0168-1699
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:•A novel chicken monitoring system for poultry industry frees farmers.•Attention mechanism-enhanced YOLO-Chicken algorithm outperforms the others.•BoT-SORT tracking algorithm precisely monitors the track of the chickens.•A graphical interface is developed to visually display tracking status. The modern poultry industry tends to scale and standardization, characterized by rapidly growing global demand for poultry products. Advancements in technology have propelled the significance of monitoring tools. In this paper, a novel real-time monitoring system for poultry industry based on attention mechanism-enhanced YOLO-Chicken and BoT-SORT tracking algorithm is proposed. The YOLO-Chicken algorithm is presented to improve the accuracy and speed of chicken detection. The experimental results indicate that YOLO-Chicken significantly improves the performance with an accuracy that exceeds 1.35 % of the original YOLOv8 model and outperforms other classical algorithms including Faster R-CNN, RetinaNet, and YOLO series in terms of accuracy and efficiency. BoT-SORT tracking algorithm is employed to monitor the track of the chickens and a graphical interface is developed to display real-time tracking status, statistics, and warning information. This system can significantly improve the efficiency of large-scale farms and free management personnel.
ISSN:0168-1699
DOI:10.1016/j.compag.2025.110640