We are delighted to announce that a collaborative research paper, featuring a COMP undergraduate student, ZHEN Zhang Alex, and our Senior Lecturer, Dr MOHAMMED Aquil Mirza, as co-authors, has received the prestigious Best Paper Award at the 2025 International Conference on Advanced Computing and Intelligent Robotics Applications (ACIRA 2025), held in Guangzhou, China, from 7- 9 November 2025.
The paper, titled "Texture-Based Classification Techniques for Sports Surface and Equipment Analysis Using Deep Learning," is the result of a cross-disciplinary effort involving students from COMP, Department of Applied Mathematics (AMA) and Department of Civil and Environmental Engineering (CEE), staff members from the Industrial Centre (IC).
Authors:
- ZHEN Zhang Alex (COMP Year 4 Student)
- LIO Kun Hon (CEE Student)
- CHAN Ka Wan (CEE Student)
- YIP Hiu Ying, (CEE Student)
- NG Hoi Yin (CEE Student)
- LUI Matthew (CEE Student)
- ZHANG Jiaming (AMA Student)
- SUN Wanxu (AMA Student)
- Dr MOHAMMED Aquil Mirza (Senior Lecturer, COMP)
- Mr LEE Kam-pui (IC)
- Dr XIAO Mingxiang (IC)
- Mr CHAN Yu-Keung (IC)
Originating as a semester project in COMP1012 in 2024 and later extended in collaboration with the IC department, the study explores the application of advanced texture-based classification techniques using deep learning (DL) architectures for analysing sports surfaces and equipment. Recognising the critical role of surface textures and material properties in sports performance, safety, and maintenance, the team evaluated the effectiveness of various deep learning models—including Convolutional Neural Networks (CNNs) and Vision Transformers—in accurately identifying and categorising surface materials and equipment textures.
This work provides a robust framework for automated quality control and predictive maintenance in sports facility management, demonstrating a practical application of advanced Artificial Intelligence (AI) in automation to enhance safety and operational efficiency. The study also addresses practical considerations such as computational efficiency, model interpretability, and deployment challenges, paving the way for real-world implementation of AI-driven surface and equipment analysis in sports technology.
