Uncertainty-Aware Fourier Ptychography (UA-FP)
Professor Edmund Lam, Dr Ni Chen and their research team from the Department of Electrical and Electronic Engineering under the Faculty of Engineering at the University of Hong Kong (HKU) have developed a novel uncertainty-aware Fourier ptychography (UA-FP) technology that significantly enhancing imaging system stability in complex real-world environments. The breakthrough has been published in Light: Science & Applications, an international journal under Nature Publishing Group.
Fourier ptychography, widely regarded as a cornerstone of computational imaging, enables wide field-of-view and high-resolution imaging with broad applications ranging from microscopy to X-ray and remote sensing. However, its practical implementation has long been hindered by misalignments, optical aberrations, and poor data quality — challenges common across computational imaging fields.
The HKU-led team's UA-FP framework innovatively incorporates uncertainty parameters into a fully differentiable computational model, enabling simultaneous system uncertainty quantification and correction and significant enhancement of imaging performance—even under suboptimal or interference-prone conditions. This advancement represents not only a breakthrough in ptychography but also a transformative development for computational imaging as a whole.
Building on the team's pioneering work in differentiable imaging since 2021, UA-FP leverages differentiable programming—the foundational principle behind deep learning—to establish an end-to-end computational framework that seamlessly integrates optical hardware, mathematical modeling, and algorithmic reconstruction. This unified approach bridges hardware and software while harmonising theory with practical implementation, fostering deeper interdisciplinary collaboration between optics and computational science. As a result, differentiable imaging has emerged as a key enabler of future innovations, not only in computational imaging but across related technological fields.
Professor Edmund Lam, corresponding author of the study, said, "By embedding uncertainties into a differentiable model, we have made Fourier ptychography practical and robust. This approach provides a blueprint for advancing many other computational imaging techniques."
Lead author Dr. Ni Chen added, "This research is the most comprehensive application of differentiable imaging to date. It shows how differentiable programming can unify optics and computation, unlocking new opportunities across science and engineering.
Link to the papers:
- Ni Chen, Yang Wu, Chao Tan, Liangcai Cao, Jun Wang, Edmund Y. Lam，《Uncertainty-Aware Fourier Ptychography》，Light: Science & Applications (2025) https://doi.org/10.1038/s41377-025-01915-w
- Ni Chen, David J. Brady, Edmund Y. Lam, Differentiable Imaging: Progress, Challenges, and Outlook, Advanced Devices & Instrumentation (2025) https://doi.org/10.34133/adi.0117
About Professor Edmund Lam
Prof. Edmund Lam is currently a Professor in the Department of Electrical and Electronic Engineering, and Associate Dean of Graduate School. Computational optics and imaging form the main focus of the work of Prof. Lam, whose broad research interests span from the design of algorithms and systems to applications especially in semiconductor manufacturing and biomedicine. He is a Fellow of several professional societies, including Optica, the Society of Photo-optical Instrumentation Engineers (SPIE), the Institute of Electrical and Electronics Engineers (IEEE), the Society for Imaging Science and Technology (IS&T), Institute of Physics (IOP), and the Hong Kong Institution of Engineers (HKIE). He is also a Founding Member of the Hong Kong Young Academy of Sciences.
About Dr Ni Chen
Dr. Ni Chen is an Honorary Associate Professor at the Department of Electrical and Electronic Engineering, The University of Hong Kong. She obtained her Ph.D. degree from the Department of Electrical and Computer Engineering at Seoul National University, Korea. She has worked in various institutions like The University of Hong Kong, Shanghai Institute of Optics and Fine Mechanics, China, Seoul National University, Korea, King Abdullah University of Science and Technology, Saudi Arabia, and the University of Arizona, USA. Her current research primarily centers on computational imaging, with a specific focus on differentiable imaging. This research aims to address the mismatch among the key components of computational imaging systems. Operating at the intersection of optics and computational science, she is dedicated to exploring the fundamental principles underlying different disciplines within computational imaging and endeavors to develop unified solutions for tackling the co - design challenges in computational imaging systems.
