skip to main content
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Real-time coherent diffraction inversion using deep generative networks

Abstract

Phase retrieval, or the process of recovering phase information in reciprocal space to reconstruct images from measured intensity alone, is the underlying basis to a variety of imaging applications including coherent diffraction imaging (CDI). Typical phase retrieval algorithms are iterative in nature, and hence, are time-consuming and computationally expensive, making real-time imaging a challenge. Furthermore, iterative phase retrieval algorithms struggle to converge to the correct solution especially in the presence of strong phase structures. In this work, we demonstrate the training and testing of CDI NN, a pair of deep deconvolutional networks trained to predict structure and phase in real space of a 2D object from its corresponding far-field diffraction intensities alone. Furthermore, once trained, CDI NN can invert a diffraction pattern to an image within a few milliseconds of compute time on a standard desktop machine, opening the door to real-time imaging.

Authors:
ORCiD logo [1]; ORCiD logo [1];  [1]
  1. Argonne National Lab. (ANL),Lemont, IL (United States)
Publication Date:
Research Org.:
Argonne National Lab. (ANL), Argonne, IL (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Basic Energy Sciences (BES) (SC-22)
OSTI Identifier:
1493761
Grant/Contract Number:  
AC02-06CH11357
Resource Type:
Accepted Manuscript
Journal Name:
Scientific Reports
Additional Journal Information:
Journal Volume: 8; Journal Issue: 1; Journal ID: ISSN 2045-2322
Publisher:
Nature Publishing Group
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING

Citation Formats

Cherukara, Mathew J., Nashed, Youssef S. G., and Harder, Ross J. Real-time coherent diffraction inversion using deep generative networks. United States: N. p., 2018. Web. doi:10.1038/s41598-018-34525-1.
Cherukara, Mathew J., Nashed, Youssef S. G., & Harder, Ross J. Real-time coherent diffraction inversion using deep generative networks. United States. doi:10.1038/s41598-018-34525-1.
Cherukara, Mathew J., Nashed, Youssef S. G., and Harder, Ross J. Thu . "Real-time coherent diffraction inversion using deep generative networks". United States. doi:10.1038/s41598-018-34525-1. https://www.osti.gov/servlets/purl/1493761.
@article{osti_1493761,
title = {Real-time coherent diffraction inversion using deep generative networks},
author = {Cherukara, Mathew J. and Nashed, Youssef S. G. and Harder, Ross J.},
abstractNote = {Phase retrieval, or the process of recovering phase information in reciprocal space to reconstruct images from measured intensity alone, is the underlying basis to a variety of imaging applications including coherent diffraction imaging (CDI). Typical phase retrieval algorithms are iterative in nature, and hence, are time-consuming and computationally expensive, making real-time imaging a challenge. Furthermore, iterative phase retrieval algorithms struggle to converge to the correct solution especially in the presence of strong phase structures. In this work, we demonstrate the training and testing of CDI NN, a pair of deep deconvolutional networks trained to predict structure and phase in real space of a 2D object from its corresponding far-field diffraction intensities alone. Furthermore, once trained, CDI NN can invert a diffraction pattern to an image within a few milliseconds of compute time on a standard desktop machine, opening the door to real-time imaging.},
doi = {10.1038/s41598-018-34525-1},
journal = {Scientific Reports},
number = 1,
volume = 8,
place = {United States},
year = {2018},
month = {11}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Citation Metrics:
Cited by: 3 works
Citation information provided by
Web of Science

Figures / Tables:

Figure 1 Figure 1: Structure of the deep generative network CDI NN. CDI NN is implemented using an architecture composed entirely of convolutional, max pooling and upsampling layers. All activations are rectified linear units (ReLU) except for the final convolutional layer which uses sigmoidal activations.

Save / Share:

Works referenced in this record:

Coherent X-ray diffraction imaging of strain at the nanoscale
journal, April 2009

  • Robinson, Ian; Harder, Ross
  • Nature Materials, Vol. 8, Issue 4
  • DOI: 10.1038/nmat2400

Deep learning
journal, May 2015

  • LeCun, Yann; Bengio, Yoshua; Hinton, Geoffrey
  • Nature, Vol. 521, Issue 7553
  • DOI: 10.1038/nature14539

Deep Learning Enhanced Mobile-Phone Microscopy
journal, March 2018


Three-dimensional imaging of strain in a single ZnO nanorod
journal, December 2009

  • Newton, Marcus C.; Leake, Steven J.; Harder, Ross
  • Nature Materials, Vol. 9, Issue 2
  • DOI: 10.1038/nmat2607

Phase retrieval from the magnitude of the Fourier transforms of nonperiodic objects
journal, January 1998

  • Miao, J.; Sayre, D.; Chapman, H. N.
  • Journal of the Optical Society of America A, Vol. 15, Issue 6
  • DOI: 10.1364/JOSAA.15.001662

Avalanching strain dynamics during the hydriding phase transformation in individual palladium nanoparticles
journal, December 2015

  • Ulvestad, A.; Welland, M. J.; Collins, S. S. E.
  • Nature Communications, Vol. 6, Issue 1
  • DOI: 10.1038/ncomms10092

Approximation capabilities of multilayer feedforward networks
journal, January 1991


Beyond crystallography: Diffractive imaging using coherent x-ray light sources
journal, April 2015


Phase recovery and holographic image reconstruction using deep learning in neural networks
journal, October 2017

  • Rivenson, Yair; Zhang, Yibo; Günaydın, Harun
  • Light: Science & Applications, Vol. 7, Issue 2
  • DOI: 10.1038/lsa.2017.141

Bragg coherent diffractive imaging of single-grain defect dynamics in polycrystalline films
journal, May 2017


Searching with iterated maps
journal, January 2007

  • Elser, V.; Rankenburg, I.; Thibault, P.
  • Proceedings of the National Academy of Sciences, Vol. 104, Issue 2
  • DOI: 10.1073/pnas.0606359104

Ultrafast Three-Dimensional X-ray Imaging of Deformation Modes in ZnO Nanocrystals
journal, January 2017


Ror2 signaling regulates Golgi structure and transport through IFT20 for tumor invasiveness
journal, January 2017


High-resolution ab initio three-dimensional x-ray diffraction microscopy
journal, January 2006

  • Chapman, Henry N.; Barty, Anton; Marchesini, Stefano
  • Journal of the Optical Society of America A, Vol. 23, Issue 5
  • DOI: 10.1364/JOSAA.23.001179

Phase retrieval algorithms: a comparison
journal, January 1982


Image reconstruction by domain-transform manifold learning
journal, March 2018

  • Zhu, Bo; Liu, Jeremiah Z.; Cauley, Stephen F.
  • Nature, Vol. 555, Issue 7697
  • DOI: 10.1038/nature25988

Imaging transient melting of a nanocrystal using an X-ray laser
journal, June 2015

  • Clark, Jesse N.; Beitra, Loren; Xiong, Gang
  • Proceedings of the National Academy of Sciences, Vol. 112, Issue 24
  • DOI: 10.1073/pnas.1417678112

Three-Dimensional Integrated X-ray Diffraction Imaging of a Native Strain in Multi-Layered WSe 2
journal, February 2018


Application of optimization technique to noncrystalline x-ray diffraction microscopy: Guided hybrid input-output method
journal, August 2007


Atomic Resolution Imaging of a Carbon Nanotube from Diffraction Intensities
journal, May 2003


Three-dimensional mapping of a deformation field inside a nanocrystal
journal, July 2006

  • Pfeifer, Mark A.; Williams, Garth J.; Vartanyants, Ivan A.
  • Nature, Vol. 442, Issue 7098
  • DOI: 10.1038/nature04867

Sparsity-based single-shot subwavelength coherent diffractive imaging
journal, April 2012

  • Szameit, A.; Shechtman, Y.; Osherovich, E.
  • Nature Materials, Vol. 11, Issue 5
  • DOI: 10.1038/nmat3289

Fully Convolutional Networks for Semantic Segmentation
journal, April 2017

  • Shelhamer, Evan; Long, Jonathan; Darrell, Trevor
  • IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 39, Issue 4
  • DOI: 10.1109/TPAMI.2016.2572683

Learning approach to optical tomography
journal, January 2015

  • Kamilov, Ulugbek S.; Papadopoulos, Ioannis N.; Shoreh, Morteza H.
  • Optica, Vol. 2, Issue 6
  • DOI: 10.1364/OPTICA.2.000517

Ultrafast Three-Dimensional Integrated Imaging of Strain in Core/Shell Semiconductor/Metal Nanostructures
journal, November 2017


Strain-relief by single dislocation loops in calcite crystals grown on self-assembled monolayers
journal, June 2016

  • Ihli, Johannes; Clark, Jesse N.; Côté, Alexander S.
  • Nature Communications, Vol. 7, Issue 1
  • DOI: 10.1038/ncomms11878

In situ strain evolution during a disconnection event in a battery nanoparticle
journal, January 2015

  • Ulvestad, Andrew; Clark, Jesse N.; Singer, Andrej
  • Physical Chemistry Chemical Physics, Vol. 17, Issue 16
  • DOI: 10.1039/C5CP00372E

Three-dimensional imaging of dislocation propagation during crystal growth and dissolution
journal, June 2015

  • Clark, Jesse N.; Ihli, Johannes; Schenk, Anna S.
  • Nature Materials, Vol. 14, Issue 8
  • DOI: 10.1038/nmat4320

X-ray image reconstruction from a diffraction pattern alone
journal, October 2003


    Figures/Tables have been extracted from DOE-funded journal article accepted manuscripts.