Professor Zhao Xiaoming’s team at Taizhou University recently published two papers— “YMAD: An Efficient Poultry Gender Classification Method Based on Feature Fusion and YOLO Model” and “YBEDI: A Lightweight and Efficient Multi-scale Feature Fusion Gender Classification Model for Modern Agriculture”—in the international agricultural journals Computers and Electronics in Agriculture (CAS Q1 TOP, IF = 8.9) and Smart Agricultural Technology (CAS Q2, IF = 5.7), respectively. Professor Zhao Xiaoming is the first author, and TU is listed as the first affiliation.

In poultry production, accurate sex identification of chicks is essential for optimizing feeding and management. Traditional techniques, such as laparoscopy and genetic testing, can determine chick sex, but they are complex, time-consuming, and require trained specialists, which limits their suitability for routine on-farm use. To address these challenges, the team established a dataset of chick gonad images and proposed a deep-learning method called YMAD (Figure 1) for chick sex classification, aiming to provide a more practical alternative to conventional approaches.

Schematic diagram of the YMAD network structure

To improve robustness under real-world conditions—particularly interference from feathers and variations in lighting—the team further integrated the YOLOv11 detection module with BEDI to develop the YBEDI model (Figure 2). This approach reduces reliance on extensive manual procedures while improving recognition accuracy. The model is highly lightweight, with only 6.24 million parameters, enabling efficient deployment on standard computers as well as edge and embedded devices. Experimental results show that YBEDI achieves 92.5% accuracy across different processor platforms, offering a strong balance of real-time performance and reliability and providing a practical solution for intelligent upgrading in commercial poultry farming.

Schematic Diagram of YBEDI Network Structure

The above research results were supported by Zhejiang Provincial Science and Technology Plan Project “Pioneer and Leading Goose + X” R&D Program (2024C03260, 2023C03197) and Taizhou City Science and Technology Plan Project (24nya19).

Paper 1 Link: https://doi.org/10.1016/j.compag.2025.111315

Paper 2 Link: https://doi.org/10.1016/j.atech.2025.101552