Generative adversarial network for super-resolution imaging through a fiber

TL;DR

This paper introduces a machine learning-based super-resolution imaging method through multimode fibers using a generative adversarial network, achieving high-quality, noise-robust images below the diffraction limit at sub-Nyquist speeds.

Contribution

It presents a novel fiber imaging approach combining compressive sensing with a data-driven GAN for improved image reconstruction without sparsity constraints.

Findings

Outperforms conventional algorithms in image quality and noise robustness

Demonstrates speckle-based super-resolution imaging below the diffraction limit

Achieves sub-Nyquist imaging speed through a multimode fiber

Abstract

A multimode fiber represents the ultimate limit in miniaturization of imaging endoscopes. Here we propose a fiber imaging approach employing compressive sensing with a data-driven machine learning framework. We implement a generative adversarial network for image reconstruction without relying on a sample sparsity constraint. The proposed method outperforms the conventional compressive imaging algorithms in terms of image quality and noise robustness. We experimentally demonstrate speckle-based imaging below the diffraction limit at a sub-Nyquist speed through a multimode fiber.