Bridging the gap between electron and optical microscopy through neural network-enabled training and imaging

M. Zaini; J. Perkins; H. Cheng; B. Gholipour

doi:10.1117/12.2633249

4 October 2022 Bridging the gap between electron and optical microscopy through neural network-enabled training and imaging

M. Zaini, J. Perkins, H. Cheng, B. Gholipour

Author Affiliations +

Proceedings Volume 12239, Unconventional Imaging and Adaptive Optics 2022; 122390K (2022) https://doi.org/10.1117/12.2633249
Event: SPIE Optical Engineering + Applications, 2022, San Diego, California, United States

Abstract

Obtaining high-resolution images using an optical microscope is critical when dealing with micro/nanoscale objects. Current techniques use high magnification objective lenses with high numerical apertures to resolve closely spaced objects at the micron/nanoscale. However, these lenses often require additional optics and have a narrow depth of field, preventing ease of use. To date, scanning electron microscopy (SEM) is used for imaging beyond the diffraction limit and has led to various breakthroughs in semiconductor physics and nanotechnology. An alternative to an SEM is using artificial intelligence (AI) to enable super-resolution techniques with correlated image sets. We utilize a convolutional neural network (CNN) and generative adversarial network (GAN) to train correlated images gathered from higher magnification SEM and lower magnification SEM, resulting in a model that enables resolving nanoscale features. We demonstrated that by training a neural network with SEM images, we are able to aid the optical microscope to image beyond the diffraction limit with a resolution closer to the SEM.

Conference Presentation

Citation Download Citation

M. Zaini, J. Perkins, H. Cheng, and B. Gholipour "Bridging the gap between electron and optical microscopy through neural network-enabled training and imaging", Proc. SPIE 12239, Unconventional Imaging and Adaptive Optics 2022, 122390K (4 October 2022); https://doi.org/10.1117/12.2633249

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available