Multimode optical fiber transmission with a deep learning network

By Babak Rahmani, D. Loterie, G. Konstantinou, Demetri Psaltis, C. Moser. Light: Science & Applications, 2018

323 citations recorded by the source corpus.

Multimode fibers (MMFs) are an example of a highly scattering medium, which scramble the coherent light propagating within them to produce seemingly random patterns. Thus, for applications such as imaging and image projection through an MMF, careful measurements of the relationship between the inputs and outputs of the fiber are required. We show, as a proof of concept, that a deep neural network can learn the input-output relationship in a 0.75 m long MMF. Specifically, we demonstrate that a deep convolutional neural network (CNN) can learn the nonlinear relationships between the amplitude of the speckle pattern (phase information lost) obtained at the output of the fiber and the phase or the amplitude at the input of the fiber. Effectively, the network performs a nonlinear inversion task. We obtained image fidelities (correlations) as high as ~98% for reconstruction and ~94% for image projection in the MMF compared with the image recovered using the full knowledge of the system transmission characterized with the complex measured matrix. We further show that the network can be trained for transfer learning, i.e., it can transmit images through the MMF, which belongs to another class not used for training/testing. A convolutional neural network (CNN) can successfully learn the nonlinear transmission characteristics of a multimode fibre thus allowing accurate image transmission and reconstruction. Propagation along a multimode fibre usually scrambles an input image, resulting in a seemingly random speckle pattern at the output. Babak Rahmani and coworkers from the Ecole Polytechnique Fédérale de Lausanne in Switzerland have now shown that a deep neural network(either a 22-layer CNN based on VGG-net technology or a 20-layer CNN based on Res-net technology) can learn the input-output relationship in a 0.75 m long piece of multimode fibre) and thus undo this scrambling. Experiments showed that both neural networks could perform highly accurate image reconstruction with an image fidelity as high as ~98% for image reconstruction and ~94% for image projection in the best case.

Subfields: Acoustics and Ultrasonics, Biomedical Engineering, Instrumentation

View Paper: https://www.semanticscholar.org/paper/9dffd776a7e0799ce80f720e51464b81fb5dd1b8

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Acoustics and Ultrasonics

In Biomedical Engineering

In Instrumentation

Citations in the corpus

Roadmap on wavefront shaping and deep imaging in complex media (2021, methodology): “Using convolutional networks, Rahmani et al [202], Borhani et al [203] and Kakkava et al [204] were able to reconstruct sparse-like input images that were scrambled upon propagation through various MMF lengths up to 1 km. Figure 30(a) shows the optical setup that is used in the data collection step.”
Inference in artificial intelligence with deep optics and photonics (2020, methodology): “The approach is also amenable to training a neural network, as demonstrated for imaging through a multimode fibre 53,54 or imaging through thin 55 or thick 56 scattering media.”