skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.


Title: Machine Learning-Based Diffractive Image Analysis with Subwavelength Resolution
Far-field analysis of small objects is severely constrained by the diffraction limit. Existing tools achieving sub-diffraction resolution often utilize point-by-point image reconstruction via scanning or labelling. Here, we present a new technique capable of fast and accurate characterization of two-dimensional structures with at least wavelength/25 theoretical resolution, based on a single far-field intensity measurement. Experimentally, we realized this technique resolving the smallest-available to us 180-nm-scale features with 845-nm laser light, reaching a resolution of wavelength/5. A comprehensive analysis of machine learning algorithms was performed to gain insight into the learning process and to understand the flow of subwavelength information through the system. Image parameterization, suitable for diffractive configurations and highly tolerant to random noise was developed. The proposed technique can be applied to new optical characterization tools with high spatial resolution, fast data acquisition and artificial intelligence, such as high-speed nanoscale metrology and quality control, and can be further developed to high-resolution spectroscopy  more » « less
Award ID(s):
2026703
PAR ID:
10227755
Author(s) / Creator(s):
; ; ; ; ;
Date Published:
Journal Name:
ACS Photonics
ISSN:
2330-4022
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Periasamy, Ammasi; So, Peter T.; König, Karsten (Ed.)
    Using the structured illumination, single pixel detection imaging technique SPatIal Frequency modulation Imaging (SPIFI), we demonstrate a cascaded Wavelength Domain and Spatial Domain (WD-SD-SPIFI) system enabling real-time, in-line, second order dispersion compensation optimization for multiphoton imaging. Enhanced resolution is demonstrated by imaging a sub-diffractive 140 nm fluorescent nanodiamond with Two Photon Excitation Fluorescence (2PEF) to measure the Point Spread Function (PSF). With a 1034 nm pulsed laser through a Numerical Aperture (NA) of 0.5, a PSF Full Width at Half Max (FWHM) of 780 nm was measured with minimal post processing analysis that only requires Fast Fourier Transforms (FFTs). 
    more » « less
  2. Featuring broadband operation and high efficiency, gallium nitride (GaN)-based radio frequency (RF) power amplifiers are key components to realize the next generation mobile network. However, to fully implement GaN high electron mobility transistors (HEMT) for such applications, it is necessary to overcome thermal reliability concerns stemming from localized extreme temperature gradients that form under high voltage and power operation. In this work, we developed a deep-ultraviolet thermoreflectance thermal imaging capability, which can potentially offer the highest spatial resolution among diffraction-limited far-field optical thermography techniques. Experiments were performed to compare device channel temperatures obtained from near-ultraviolet and deep-ultraviolet wavelength illumination sources for the proof of concept of the new characterization method. Deep-ultraviolet thermoreflectance imaging will facilitate the study of device self-heating within transistors based on GaN and emerging ultra-wide bandgap semiconductors (e.g., β-Ga2O3, AlxGa1-xN, and diamond) subjected to simultaneous extreme electric field and heat flux conditions. 
    more » « less
  3. We review the recent advances in thermal characterization of micro/nanoscale electronic, optoelectronic, thermal devices based on thermoreflectance imaging. Thermoreflectance imaging is a non-invasive optical technique that can visualize surface thermal response of devices and integrated circuits (IC). Recent advances of the technique have enabled high-resolution, ultra-fast transient thermal imaging with 800 ps temporal resolution. Using visible or UV illumination, spatial resolution of about 200-250 nm can be achieved. Many IC substrates, e.g. Si, GaAs, are transparent to near IR illumination in 1-1.5 μm wavelength range. Through-substrate thermal imaging of flip-chip bonded ICs with micron spatial resolution has been demonstrated. We provide key examples of various devices characterized by the technique such as CMOS ICs, GaN HEMT, nanowire transistors, thin-film solar cells, and micro-thermal cloaking devices. In addition to the validation of electrothermal models, material and fabrication defects can be identified. Finally we discuss the advantages/limitations, and perspective of thermoreflectance imaging technique. 
    more » « less
  4. From engineering improved device performance to unraveling the breakdown of classical heat transfer laws, far-field optical temperature mapping with nanoscale spatial resolution would benefit diverse areas. However, these attributes are traditionally in opposition because conventional far-field optical temperature mapping techniques are inherently diffraction limited. Optical super-resolution imaging techniques revolutionized biological imaging, but such approaches have yet to be applied to thermometry. Here, we demonstrate a super-resolution nanothermometry technique based on highly doped upconverting nanoparticles (UCNPs) that enable stimulated emission depletion (STED) super-resolution imaging. We identify a ratiometric thermometry signal and maintain imaging resolution better than ~120 nm for the relevant spectral bands. We also form self-assembled UCNP monolayers and multilayers and implement a detection scheme with scan times >0.25 μm2/min. We further show that STED nanothermometry reveals a temperature gradient across a joule-heated microstructure that is undetectable with diffraction limited thermometry, indicating the potential of this technique to uncover local temperature variation in wide-ranging practical applications. 
    more » « less
  5. Single-shot 3D optical microscopy that can capture high-resolution information over a large volume has broad applications in biology. Existing 3D imaging methods using point-spread-function (PSF) engineering often have limited depth of field (DOF) or require custom and often complex design of phase masks. We propose a new, to the best of our knowledge, PSF approach that is easy to implement and offers a large DOF. The PSF appears to be axially V-shaped, engineered by replacing the conventional tube lens with a pair of axicon lenses behind the objective lens of a wide-field microscope. The 3D information can be reconstructed from a single-shot image using a deep neural network. Simulations in a 10× magnification wide-field microscope show the V-shaped PSF offers excellent 3D resolution (<2.5 µm lateral and ∼15 µm axial) over a ∼350 µm DOF at a 550 nm wavelength. Compared to other popular PSFs designed for 3D imaging, the V-shaped PSF is simple to deploy and provides high 3D reconstruction quality over an extended DOF. 
    more » « less