Real-time coherent diffraction inversion using deep generative networks

TL;DR

This paper introduces a deep learning approach using neural networks to perform real-time phase retrieval in coherent diffraction imaging, significantly speeding up the reconstruction process compared to traditional iterative algorithms.

Contribution

The authors develop and demonstrate a deep neural network method that rapidly reconstructs images from diffraction patterns, enabling real-time imaging in CDI.

Findings

Reconstruction time reduced to milliseconds on standard hardware.

Neural networks outperform iterative algorithms in speed and robustness.

Potential for real-time imaging applications in various scientific fields.

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, precluding real-time imaging. 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. 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.