• Corpus ID: 49559835

prDeep: Robust Phase Retrieval with a Flexible Deep Network

@inproceedings{Metzler2018prDeepRP,
  title={prDeep: Robust Phase Retrieval with a Flexible Deep Network},
  author={Christopher A. Metzler and Philip Schniter and Ashok Veeraraghavan and Richard Baraniuk},
  booktitle={International Conference on Machine Learning},
  year={2018}
}
Phase retrieval algorithms have become an important component in many modern computational imaging systems. For instance, in the context of ptychography and speckle correlation imaging, they enable imaging past the diffraction limit and through scattering media, respectively. Unfortunately, traditional phase retrieval algorithms struggle in the presence of noise. Progress has been made recently on more robust algorithms using signal priors, but at the expense of limiting the range of supported… 
[PDF] Semantic Reader
114 Citations
Highly Influential Citations
15
Background Citations
49
Methods Citations
54
Results Citations
5

Figures and Tables from this paper

114 Citations

prDeep: Robust Phase Retrieval with Flexible Deep Neural Networks

This work uses convolutional neural networks, a powerful tool from machine learning, to regularize phase retrieval problems and improve recovery performance and test the new algorithm, prDeep, in simulation and demonstrate that it is robust to noise, can handle a variety system models, and operates fast enough for high-resolution applications.

When deep denoising meets iterative phase retrieval

This work combines iterative methods from phase retrieval with image statistics from deep denoisers, via regularization-by-denoising, and paves the way for hybrid imaging methods that integrate machine-learned constraints in conventional algorithms.

Compressed Sensing-Based Robust Phase Retrieval via Deep Generative Priors

This article proposes a framework to regularize the highly ill-posed and non-linear phase retrieval problem through deep generative priors by simply using gradient descent algorithm and modify the proposed algorithm to allow the generative model to explore solutions outside its range, leading to improved performance.

Real-time coherent diffraction inversion using deep generative networks

This work demonstrates 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, opening the door to real-time imaging.

Phase Retrieval: From Computational Imaging to Machine Learning: A tutorial

This tutorial reviews phase retrieval under a unifying framework that encompasses classical and machine learning methods and focuses on three key elements: applications, an overview of recent reconstruction algorithms, and the latest theoretical results.

Class-Specific Blind Deconvolutional Phase Retrieval Under a Generative Prior

The proposed recovery algorithm strives to find a sharp image and a blur kernel in the range of the respective pre-generators that explain the forward measurement model and is able to reconstruct quality image estimates.

Deep Iterative Reconstruction for Phase Retrieval

This work develops a phase retrieval algorithm that utilizes two DNNs together with the model-based HIO method that not only achieves state-of-the-art reconstruction performance but also is more robust to different initialization and noise levels.

Deep Iterative Phase Retrieval for Ptychography

This work proposes an augmentation of existing iterative phase retrieval algorithms with a neural network designed for refining the result of each iteration, which adapt and extend a recently proposed architecture from the speech processing field.

Non-Iterative Phase Retrieval With Cascaded Neural Networks

A deep neural network cascade that reconstructs the image successively on different resolutions from its non-oversampled Fourier magnitude and yields improved performance over other non-iterative methods and optimization-based methods.

Physics-driven Deep Neural Network for Fourier Phase Retrieval

Since the process is non-iterative and the network is end-to-end trained, it is much faster and more accurate than the traditional physics-driven PR approaches, and demonstrates the superiority of the proposed PPRNet over the traditional PR methods.
...

References

SHOWING 1-10 OF 58 REFERENCES

prDeep: Robust Phase Retrieval with Flexible Deep Neural Networks

This work uses convolutional neural networks, a powerful tool from machine learning, to regularize phase retrieval problems and improve recovery performance and test the new algorithm, prDeep, in simulation and demonstrate that it is robust to noise, can handle a variety system models, and operates fast enough for high-resolution applications.

BM3D-PRGAMP: Compressive phase retrieval based on BM3D denoising

This paper uses denoisers to impose elaborate and accurate models in order to perform inference on generalized linear systems and demonstrates recovery performance equivalent to existing techniques using fewer than half as many measurements.

Oversampling smoothness: an effective algorithm for phase retrieval of noisy diffraction intensities.

Both numerical simulations with Poisson noise and experimental data from a biological cell indicate that OSS consistently outperforms the HIO, ER-HIO and noise robust (NR)-HIO algorithms at all noise levels in terms of accuracy and consistency of the reconstructions.

PhasePack: A phase retrieval library

The purpose of PhasePack is to create a common software interface for a wide range of phase retrieval algorithms and to provide a common testbed using both synthetic data and empirical imaging datasets.

DOLPHIn—Dictionary Learning for Phase Retrieval

This work proposes a new algorithm to learn a dictionary for reconstructing and sparsely encoding signals from measurements without phase, and jointly reconstructs the unknown signal and learns a dictionary such that each patch of the estimated image can be sparsely represented.

Phase retrieval from noisy data based on sparse approximation of object phase and amplitude

A variational approach to reconstruction of phase and amplitude of a complex-valued object from Poissonian intensity observations is developed and demonstrates a valuable advantage for heavily noisy observations corresponding to a short exposure time in optical experiments.

Ptychnet: CNN based fourier ptychography

This paper proposes a new reconstruction algorithm that is based on convolutional neural networks and demonstrates its advantages in terms of speed and performance.

Unrolled Optimization with Deep Priors

This paper presents unrolled optimization with deep priors, a principled framework for infusing knowledge of the image formation into deep networks that solve inverse problems in imaging, inspired by classical iterative methods.

Phase recovery and holographic image reconstruction using deep learning in neural networks

It is demonstrated that a neural network can learn to perform phase recovery and holographic image reconstruction after appropriate training, and this deep learning-based approach provides an entirely new framework to conduct holographic imaging by rapidly eliminating twin-image and self-interference-related spatial artifacts.

Plug-and-Play priors for model based reconstruction

This paper demonstrates with some simple examples how Plug-and-Play priors can be used to mix and match a wide variety of existing denoising models with a tomographic forward model, thus greatly expanding the range of possible problem solutions.
...