Abstract:Single-fiber imaging technology offers unique insights for research and inspection in difficult-to-reach and narrow spaces. In particular, ultra-compact multimode fiber (MMF) imaging, has received increasing interest over the past decade. However, MMF imaging will be seriously distorted when subjected to dynamic perturbations due to time-varying mode coupling, and the imaging of space objects via Gaussian beam will be relatively degraded at the edge due to insufficient contrast. Here, a robust super-resolution image transmission method based on a ring core fiber (RCF) with orbital angular momentum (OAM) is proposed and experimentally demonstrated. The OAM modes propagating in the RCF form a series of weakly coupled mode groups, making the imaging system robust to external perturbations. In addition, a spiral phase plate is used as a vortex filter to produce OAM for edge enhancement, thus improving the image resolution. Furthermore, a few-shot U-Transformer neural network is proposed to enhance the resilience of the developed RCF-OAM imaging system against environmental perturbations. Finally, the developed RCF-OAM imaging system achieves biological image transmission, demonstrating the practicality of the scheme. This pioneering RCF OAM image transmission system may have broad applications, potentially revolutionizing fields such as biological imaging and industrial non-destructive testing. A ring core fiber-based image transmission method with orbital angular momentum is proposed. This method exhibits robustness to external perturbations due to the weakly coupled mode groups, achieves high resolution through OAM imaging, and introduces a network with better generalization capabilities. The excellent performance of the RCF-OAM system is demonstrated in the imaging of biological slices.image

All-optical Image Transmission Through a Dynamically Perturbed Multimode Fiber and a Ring-Core Fiber Using Diffractive Deep Neural Networks.

Diffractive Deep Neural Network for Optical Orbital Angular Momentum Multiplexing and Demultiplexing

All-Optical Signal Processing of Vortex Beams with Diffractive Deep Neural Networks

Efficient Reference-Less Transmission Matrix Retrieval for a Multimode Fiber Using Fast Fourier Transform

Orbital Angular Momentum Mode Demodulation with Neural Network-Assisted Coherent Nanophotonic Circuits

Robust Super-Resolution Image Transmission Based on a Ring Core Fiber with Orbital Angular Momentum

Spatially Variant Deblur and Image Enhancement in a Single Multimode Fiber Imaged by Deep Learning

Dynamic perturbation mitigation via polarization difference neural network for high-fidelity ring core fiber image transmission

Ring Core Fiber Based Robust Super-Resolution Imaging with Orbital Angular Momentum

Learning to transmit images through optical speckle of a multimode fiber with high fidelity

Robust Super-Resolution Imaging Based on a Ring Core Fiber with Orbital Angular Momentum

Multimode optical fiber transmission with a deep learning network

Seeing Through Multimode Fibers with Physics-Assisted Deep Learning

Optical mode manipulation using deep spatial diffractive neural networks

Learning-enabled recovering scattered data from twisted light transmitted through a long standard multimode fiber

Diffractive Neural Network on a 3D Photonic Device for Spatial Mode Bases Mapping

Deep Learning Image Transmission Through a Multimode Fiber Based on a Small Training Dataset.

Single-ended Recovery of Optical fiber Transmission Matrices using Neural Networks

In Situ Optical Backpropagation Training of Diffractive Optical Neural Networks

All-Optical Orbital Angular Momentum Encoded Diffractive Networks

Optical authentication scheme based on all-optical neural network