Prof. George Barbastathis
Massachusetts Institute of Technology, USA
Title: Darker, smaller, faster:
using machine learning and optical physics for extreme imaging
Abstract: For the past decade, my group has been working on machine learning for computational imaging and, more specifically, supervised training of regularizers for ill-posed and ill-conditioned inverse problems. The notion of data-driven regularization has been around since at least the invention of learned sparse codes, or “dictionaries,” by Olshausen and Field in 1996. Learning the code through a deep neural network, as pointed out in 2010 by Gregor and Lecun, narrows the regularizer on-demand. Thus, improvements should be expected in terms of resilience to noise and incomplete measurements. Indeed, circa 2017~18, Unser’s group at EPFL devised a cascaded neural proximal gradient scheme for linear tomography with reduced exposure dose; Ozcan’s group at UCLA showed that a low-NA microscope objective can capture features of comparable fidelity as high-NA; and my group at MIT discovered that a Gerchberg-Saxton-Fienup module followed by a deep neural network can retrieve the phase of the complex field robustly from highly noisy intensity measurements.
Using neural networks for imaging in various contexts is now widespread. What have we learned from these deployments? It is certainly true that neural networks capture priors effectively. It is less clear how optical physics, to the degree that it is involved in machine training and if at all, influences the reconstruction accuracy. One clear accomplishment from our work was to image “darker,” i.e. with extremely few photons, “smaller,” i.e. with finer spatial sampling and “faster” i.e. with finer temporal sampling than classical approaches to resolution would have allowed; and to do this with almost quantifiable accuracy bounds (though more work is certainly needed on this front.) I will further discuss recent work by my group on image-based parameter estimation for dynamical phenomena that are similarly photon-limited and severely undersampled in space and time.
Acknowledgment: Recent work was funded by the US Air Force Office of Scientific Research, Takeda Pharmaceuticals, and Fujikura Ltd. The author acknowledges working with Claude, an artificial intelligence; the author has reviewed the AI’s work and assumes full responsibility for the results.
Biodata:
George Barbastathis received the Diploma in Electrical and Computer Engineering in 1993 from the National Technical University of Athens (Εθνικό Μετσόβιο Πολυτεχνείο) and the MSc and PhD degrees in Electrical Engineering in 1994 and 1997, respectively, from the California Institute of Technology (Caltech). After post-doctoral work at the University of Illinois at Urbana-Champaign, he joined the faculty at MIT in 1999, where he is now the Ralph E. and Eloise F. Cross Professor of Manufacturing and Professor of Mechanical Engineering. He has held sabbatical appointments at Harvard University and the University of Michigan – Shanghai Jiao Tong University Joint Institute (密西根交大学院), and has been one of the longest-serving Principal Investigators with the Singapore-MIT Alliance for Research and Technology (SMART). His research interests are in machine learning and optimization for computational imaging and inverse problems; and in optical physics, including statistical optics, scattering theory, and artificial optical materials and interfaces. He is presently leading the AFOSR MURI “Searching for what’s new: the systematic development of dynamic x-ray microscopy,” whose goal is to combine dynamical principles and machine learning for imaging extreme phenomena with sub-microsecond and sub-nanometer dynamics. He is member of the Institute of Electrical and Electronics Engineering (IEEE) and the American Mathematical Society (AMS), a Fellow of the Optical Society of America (OSA), which was recently rebranded as Optica (not to be confused with the homonymous journal published by the same society), and a Fellow of the Society for Photo Instrumentation Engineering (SPIE). He has served as Associate Editor for the Journal of the Optical Society of America A and the journal Optica (not to be confused with the professional society publishing it), as a committee member, chair and co-chair of numerous conferences, and in several co-founding and consulting capacities for incumbent and start-up companies, as well as law firms. In his free time, he enjoys arthouse movies, urban art and design, and he aspires to become more fluent in Thai and Mandarin Chinese.
Prof. Ignacio Moreno
Prof. Ignacio Moreno, Universidad Miguel Hernández de Elche, Spain
Title: Spectral imaging with liquid-crystal modulators
Abstract: This talk will introduce liquid-crystal (LC) optical modulators. They have become key components for the precise control of the intensity, phase and polarization of light. The presentation will focus on their use in spectral imaging. The wavelength dependance of the LC retardance is exploited to produce tunable birefringent spectral filters to selectively acquire images in different spectral bands. Hyperspectral imaging combined with polarimetry provides a powerful tool for the inspection of birefringent samples. The use of high-resolution pixelated LC spatial light modulators will be shown, exploited to create precise spatial spectral and polarization optical elements.
TecnOPTO-Lab UMH. https://tecnopto.umh.es/
Biodata: