Learning to see through multimode fibers

Abstract

We use Deep Neural Networks (DNNs) to classify and reconstruct a large database of handwritten digits from the intensity of the speckle patterns that result after the images propagated through multimode fibers (MMF). Images transmitted through fibers with up to 1km length were recovered. The ability of the network to recognize the input degraded with fiber length but the performance could be enhanced if the neural networks were trained to first reconstruct the image and then classify it rather than classify it directly from the speckle intensity.

References in corpus (6)

• Space Division Multiplexing in Optical Fibres
• Deep Learning with Coherent Nanophotonic Circuits
• Neuromorphic Silicon Photonic Networks
• Deep Learning Microscopy
• Controlling Light Through Optical Disordered Media : Transmission Matrix Approach
• Real-time resilient focusing through a bending multimode fiber

Cited by in corpus (50)

• Deep speckle correlation: a deep learning approach towards scalable imaging through scattering media
• Deep learning in nano-photonics: inverse design and beyond
• Transmission of natural scene images through a multimode fibre
• Roadmap on Wavefront Shaping and deep imaging in complex media
• Terahertz Pulse Shaping Using Diffractive Surfaces
• Learning to decompose the modes in few-mode fibers with deep convolutional neural network
• Phase Imaging with Computational Specificity (PICS) for measuring dry mass changes in sub-cellular compartments
• Artificial Neural Network with Physical Dynamic Metasurface Layer for Optimal Sensing
• Controlling light propagation in multimode fibers for imaging, spectroscopy and beyond
• Compressively sampling the optical transmission matrix of a multimode fibre
• Light scattering control in transmission and reflection with neural networks
• Image transmission through a flexible multimode fiber by deep learning
• Intelligent Meta-Imagers: From Compressed to Learned Sensing
• Learning-based real-time method to looking through scattering medium beyond the memory effect
• Wavefront shaping in multimode fibers by transmission matrix engineering
• Observing distant objects with a multimode fibre-based holographic endoscope
• Seeing through multimode fibers with real-valued intensity transmission matrices
• Controlling spatiotemporal nonlinearities in multimode fibers with deep neural networks
• Deep learning enabled design of complex transmission matrices for universal optical components
• Intensity-only Mode Decomposition on Multimode Fibers using a Densely Connected Convolutional Network
• High-Resolution Limited-Angle Phase Tomography of Dense Layered Objects Using Deep Neural Networks
• Deep Learning Enabled Real Time Speckle Recognition and Hyperspectral Imaging using a Multimode Fiber Array
• Displacement-agnostic coherent imaging through scatter with an interpretable deep neural network
• Deep learning enabled strategies for modelling of complex aperiodic plasmonic metasurfaces of arbitrary size
• Far-field subwavelength acoustic imaging by deep learning
• Confocal 3D reflectance imaging through multimode fibers without wavefront shaping
• How to build the optical inverse of a multimode fibre
• Spatial images from temporal data
• Deep Hybrid Scattering Image Learning
• Deep learning enabled superfast and accurate M^2 evaluation for fiber beams
• Ptycho-endoscopy on a lensless ultrathin fiber bundle tip
• Spectral pre-modulation of training examples enhances the spatial resolution of the Phase Extraction Neural Network (PhENN)
• Referenceless characterisation of complex media using physics-informed neural networks
• Recognizing three-dimensional phase images with deep learning
• Generative adversarial network for super-resolution imaging through a fiber
• Spatiotemporal focusing through a multimode fiber via time-domain wavefront shaping
• Optical transmission matrix measurement sampled on a dense hexagonal lattice
• Spectral Shaping in a Multimode Fiber by All-fiber Modulation
• Deep learning-assisted imaging through stationary scattering media
• Reciprocity-induced symmetry in the round-trip transmission through complex systems
• Speckle-correlation imaging through a kaleidoscopic multimode fiber
• Tutorial: How to build and control an all-fiber wavefront modulator using mechanical perturbations
• Focal field analysis of highly multi-mode fiber beams based on modal decomposition
• Inverse Design of All-dielectric Parallel-Plane Mirror Resonator
• Polarimetric compressed sensing with hollow, self-assembled diffractive films
• Diffractive Interconnects: All-Optical Permutation Operation Using Diffractive Networks
• All-fiber microendoscopic polarization sensing at single-photon level aided by deep-learning
• Spectral splitting and concentration of broadband light using neural networks
• Focal-plane wavefront sensing with moderately broadband light using a short multi-mode fiber
• Programming multi-level quantum gates in disordered computing reservoirs via machine learning and TensorFlow