Huiyu Zhou
University of Leicester, UK

Dr. Huiyu Zhou received a Bachelor of Engineering degree in Radio Technology from Huazhong University of Science and Technology of China, and a Master of Science degree in Biomedical Engineering from University of Dundee of United Kingdom, respectively. He was awarded a Doctor of Philosophy degree in Computer Vision from Heriot-Watt University, Edinburgh, United Kingdom. Dr. Zhou currently is a full Professor at School of Computing and Mathematical Sciences, University of Leicester, United Kingdom. He has published over 600 peer-reviewed papers in the field. His research work has been or is being supported by UK EPSRC, ESRC, AHRC, MRC, EU, Innovate UK, Royal Society, British Heart Foundation, Leverhulme Trust, Puffin Trust, Alzheimer’s Research UK, Invest NI and industry.
Speech Title: TBA...
Abstract: TBA...

Ata Jahangir Moshayedi
Ata Jahangir Moshayedi DGUT-CNAM Institute, Dongguan University of Technology, China, IEEE Senior Member, ACM Member

Dr. Ata Jahangir Moshayedi is an Associate Professor at DGUT-CNAM Institute of Dongguan University of Technology, China. He holds a Ph.D. in Electronic Science from Savitribai Phule Pune University (formerly the University of Pune), India. He is a Senior Member of IEEE, a Professional Member of ACM, and a Life Member of the Instrument Society of India.Dr. Moshayedi has published over 100 research papers, authored 4 books, contributed to 4 book chapters, and holds 2 patents and 16 copyrights. He actively serves as a technical program committee member and session chair for numerous international conferences.
His current research focuses on robotics, particularly the development of autonomous guided vehicles (AGVs) for applications such as smart farming, food delivery and elderly care.
Speech Title: From Pipe Defects to Intelligent Robots AI-driven pipeline inspection From theory to real-world deployment
Abstract: Pipeline systems are essential components of modern industrial infrastructure, enabling the transportation of oil, gas, and water across large-scale networks. However, these systems are highly susceptible to various structural and operational defects, including cracks, corrosion, leakage, and deformation, which can result in significant environmental damage, safety hazards, and economic losses. This keynote presents an integrated research progression that spans from pipeline defect characterization to the development of intelligent inspection and autonomous robotic systems.
The presentation begins with a comprehensive analysis of pipeline defect types, their underlying causes, and the key challenges associated with inspection in real-world environments. This analysis highlights the limitations of conventional manual and sensor-based inspection techniques, particularly in complex, hazardous, and inaccessible conditions. To overcome these challenges, a deep learning-based defect detection framework is introduced using the YOLO model, enhanced with attention mechanisms to improve feature representation and achieve high-accuracy real-time multi-class defect detection.
Furthermore, a mobile-based system, the Handy Pipe Defect Recognizer (HPD), is developed to provide portable and real-time defect classification for field applications. In addition, a pipe inspection robot is proposed, specifically designed for operation in confined and hazardous pipeline environments, integrating vision-based perception with adaptive mobility. Finally, the keynote unifies these contributions into a comprehensive smart infrastructure framework that combines artificial intelligence, mobile computing, and robotics, enabling autonomous, efficient, and scalable pipeline monitoring for next-generation industrial systems.

Azhar Imran Mudassir
Beijing University of Technology, China, IEEE Senior Member

Dr. Azhar Imran is an Associate Professor at the Department of Computer Science, Beijing University of Technology, China, with over 13 years of international research and academic experience. He specializes in Artificial Intelligence, Machine Learning, Data Science, and Image Processing, with research interests spanning intelligent systems, computational intelligence, computer vision, pattern recognition, and data-driven modeling. His work focuses on developing robust, scalable, and interpretable AI solutions applicable across diverse domains. He has published more than 100 research articles in high-impact journals, including IEEE Transactions on Neural Networks and Learning Systems, IEEE Transactions on Image Processing, and Pattern Recognition, accumulating over 4000 citations. He has been invited as a keynote speaker at prominent international conferences, including ICBDDM-24, ICCTEC-25, CCET-25, CVIT-24, ICDSP-24, and CVIT-23, highlighting his global recognition in AI and machine vision. As a Senior Member of IEEE, he serves on the editorial boards of multiple high-impact journals, including IEEE Access, MDPI Cancers, Applied Sciences, Springer Visual Computer, Taylor & Francis Biomedical Imaging & Visualization, and Multimedia Tools & Applications, and has acted as a guest editor for IEEE Transactions on Industrial Informatics and IEEE Sensors Journal. His contributions have been recognized with several prestigious awards, including the Young Marconi Award, Prime Minister's Science and Technology Award, the Embassy Honored Award from the Pakistan Embassy in Beijing, and the Outstanding Researcher and Excellence in Teaching Awards from Beijing University of Technology. He is also recognized as a ScholarGPS top 1% scientist worldwide. Through his interdisciplinary research, Dr. Imran continues to advance intelligent, interpretable, and socially impactful AI systems, shaping the next generation of computational and vision technologies.
Speech Title:Hybrid Deep Learning and Computational Optics for Explainable Medical and Machine Vision Systems
Abstract: The convergence of computational optics and machine vision is enabling a paradigm shift in intelligent imaging systems. In this keynote, I present our research contributions on hybrid deep learning frameworks that integrate image formation principles with data-driven representation learning to achieve robust, scalable, and interpretable visual understanding. Our work focuses on medical image analysis and ECG image-based multi-class diagnosis, where EfficientNet-based and convolutional architectures are systematically designed to handle complex classification tasks under real-world constraints. Key publications informing this keynote include “Automated Multi-Class ECG Diagnosis from Image Representations Using EfficientNet and Grad-CAM” and other recent work in medical image feature learning.
A distinguishing contribution of this research lies in embedding explainability into deep models through attention mechanisms and gradient-based visualization techniques, such as Grad-CAM, allowing transparent and clinically meaningful interpretation of predictions. To address challenges including noise, domain variability, and limited annotated data, we incorporate computational imaging priors alongside transfer learning and domain adaptation strategies, significantly improving model generalization and stability.
Extensive experimental evaluations demonstrate that the proposed hybrid frameworks consistently outperform conventional deep learning approaches in terms of accuracy, robustness, and interpretability across diverse datasets. Looking ahead, the keynote highlights emerging directions, including multimodal learning, transformer-based vision architectures, and explainable AI, aiming to establish trustworthy next-generation machine vision systems. These contributions underscore the critical role of unifying computational imaging and machine learning to advance applications in healthcare diagnostics, intelligent surveillance, and large-scale visual analytics.

TBA......