Recognizing three-dimensional phase images with deep learning

TL;DR

This paper introduces a deep learning-based speckle 3D reconstruction network (STRN) that recognizes phase objects behind scattering media from single-shot speckle patterns, overcoming the memory effect limitations for biomedical and astronomical imaging.

Contribution

The paper presents a novel deep learning model (STRN) capable of depth-resolved phase imaging through scattering media without relying on the memory effect.

Findings

High-fidelity 3D phase recognition behind scattering media

Single-shot, reference-free phase imaging

Potential applications in biomedical tomography and endoscopy

Abstract

Optical phase contains key information for biomedical and astronomical imaging. However, it is often obscured by layers of heterogeneous and scattering media, which render optical phase imaging at different depths an utmost challenge. Limited by the memory effect, current methods for phase imaging in strong scattering media are inapplicable to retrieving phases at different depths. To address this challenge, we developed a speckle three-dimensional reconstruction network (STRN) to recognize phase objects behind scattering media, which circumvents the limitations of memory effect. From the single-shot, reference-free and scanning-free speckle pattern input, STRN distinguishes depth-resolving quantitative phase information with high fidelity. Our results promise broad applications in biomedical tomography and endoscopy.