Published on Fri Feb 28 2020

Class-Specific Blind Deconvolutional Phase Retrieval Under a Generative Prior

Fahad Shamshad, Ali Ahmed

The problem arises in various imaging modalities such as Fourier ptychography, X-ray crystallography, and in visible light communication. The proposed recovery algorithm strives to find a sharp image and a blur kernel in the range of the respective pre-generators.

0
0
0
Abstract

In this paper, we consider the highly ill-posed problem of jointly recovering two real-valued signals from the phaseless measurements of their circular convolution. The problem arises in various imaging modalities such as Fourier ptychography, X-ray crystallography, and in visible light communication. We propose to solve this inverse problem using alternating gradient descent algorithm under two pretrained deep generative networks as priors; one is trained on sharp images and the other on blur kernels. The proposed recovery algorithm strives to find a sharp image and a blur kernel in the range of the respective pre-generators that \textit{best} explain the forward measurement model. In doing so, we are able to reconstruct quality image estimates. Moreover, the numerics show that the proposed approach performs well on the challenging measurement models that reflect the physically realizable imaging systems and is also robust to noise

Thu Mar 01 2018
Machine Learning
prDeep: Robust Phase Retrieval with a Flexible Deep Network
PrDeep is a new phase retrieval algorithm that is both robust and broadly applicable. It can handle a variety of system models. A MatConvNet implementation of prDeep is available.
0
0
0
Tue Mar 03 2020
Machine Learning
When deep denoising meets iterative phase retrieval
Conventional algorithms for retrieving the phase suffer when noise is present but display global convergence when given clean data. Recovering a signal from its Fourier intensity underlies many important applications, including lensless imaging.
1
5
8
Mon Dec 14 2020
Machine Learning
Phase Retrieval with Holography and Untrained Priors: Tackling the Challenges of Low-Photon Nanoscale Imaging
Holographic phase retrieval is the inverse problem of recovering a signal from Fourier measurements. Holographic CDI is crucial at the nanoscale, where imaging specimens such as viruses, viruses, and crystals require low-photon measurements. This data is highly polluted by Poisson shot noise and often lacks low-frequency content.
0
0
0
Fri Aug 17 2018
Machine Learning
Robust Compressive Phase Retrieval via Deep Generative Priors
This paper proposes a new framework to regularize the highly ill-posed and non-linear phase retrieval problem through deep generative priors. We experimentally show effectiveness of proposed algorithm for random Gaussian measurements and Fourier friendly measurements.
0
0
0
Thu Jul 16 2020
Machine Learning
DeepInit Phase Retrieval
This paper shows how data-driven deep generative models can be utilized to solve challenging phase retrieval problems. Classical iterative algorithms are typically known to work well if initialized close to the optimum but otherwise suffer from non-convexity and often get stuck in local minima.
0
0
0
Tue Apr 14 2020
Machine Learning
On the interplay between physical and content priors in deep learning for computational imaging
Deep learning has been applied extensively in many computational imaging problems. Two important questions remain largely unanswered. How well can the trained neural network generalize to objects very different from the ones in training? This is particularly important in practice, since large-scale annotated examples are often not available during training.
0
0
0