Non-Iterative Phase Retrieval With Cascaded Neural Networks

TL;DR

This paper introduces a deep neural network cascade for non-iterative Fourier phase retrieval, enabling high-quality image reconstruction from non-oversampled magnitude data across multiple datasets.

Contribution

It presents a novel cascaded neural network approach that improves upon existing non-iterative and optimization-based phase retrieval methods for non-oversampled data.

Findings

Outperforms other non-iterative methods

Achieves better reconstruction quality than optimization algorithms

Effective across diverse datasets

Abstract

Fourier phase retrieval is the problem of reconstructing a signal given only the magnitude of its Fourier transformation. Optimization-based approaches, like the well-established Gerchberg-Saxton or the hybrid input output algorithm, struggle at reconstructing images from magnitudes that are not oversampled. This motivates the application of learned methods, which allow reconstruction from non-oversampled magnitude measurements after a learning phase. In this paper, we want to push the limits of these learned methods by means of a deep neural network cascade that reconstructs the image successively on different resolutions from its non-oversampled Fourier magnitude. We evaluate our method on four different datasets (MNIST, EMNIST, Fashion-MNIST, and KMNIST) and demonstrate that it yields improved performance over other non-iterative methods and optimization-based methods.