Deep learning the high variability and randomness inside multimode fibres

Multimode fibers (MMF) are remarkable high-capacity information channels. However, the MMF transmission is highly sensitive to external perturbations and environmental changes. Here, we show the successful binary image transmission using deep learning through a single MMF subject to dynamic shape variations. As a proof-of-concept experiment, we find that a convolutional neural network has excellent generalization capability with various MMF transmission states to accurately predict unknown information at the other end of the MMF at any of these states. Our results demonstrate that deep learning is a promising solution to address the high variability and randomness challenge of MMF based information channels. This deep-learning approach is the starting point of developing future high-capacity MMF optical systems and devices and is applicable to optical systems concerning other diffusing media.

[1]  L. Nelson,et al.  Space-division multiplexing in optical fibres , 2013, Nature Photonics.

[2]  Stuart,et al.  Dispersive multiplexing in multimode optical fiber , 2000, Science.

[3]  Daniel A. Nolan,et al.  Self-organized instability in graded-index multimode fibres , 2016 .

[4]  Eero P. Simoncelli,et al.  Image quality assessment: from error visibility to structural similarity , 2004, IEEE Transactions on Image Processing.

[5]  Moonseok Kim,et al.  Scanner-free and wide-field endoscopic imaging by using a single multimode optical fiber. , 2012, Physical review letters.

[6]  Xin Yuan,et al.  Parallel lensless compressive imaging via deep convolutional neural networks. , 2018, Optics express.

[7]  Florent Krzakala,et al.  Reference-less measurement of the transmission matrix of a highly scattering material using a DMD and phase retrieval techniques. , 2015, Optics express.

[8]  L. Deng,et al.  The MNIST Database of Handwritten Digit Images for Machine Learning Research [Best of the Web] , 2012, IEEE Signal Processing Magazine.

[9]  Mario Krenn,et al.  Active learning machine learns to create new quantum experiments , 2017, Proceedings of the National Academy of Sciences.

[10]  Tomer Michaeli,et al.  Deep-STORM: super-resolution single-molecule microscopy by deep learning , 2018, 1801.09631.

[11]  R. Olshansky,et al.  Mode Coupling Effects in Graded-index Optical Fibers. , 1975, Applied optics.

[12]  George Barbastathis,et al.  Imaging through glass diffusers using densely connected convolutional networks , 2017, Optica.

[13]  Kin Seng Chiang,et al.  Microbend-induced mode coupling in a graded-index multimode fiber. , 2005, Applied optics.

[14]  Lei Tian,et al.  Deep speckle correlation: a deep learning approach toward scalable imaging through scattering media , 2018, Optica.

[15]  Hao Wang,et al.  Exploit imaging through opaque wall via deep learning , 2017, ArXiv.

[16]  Shanhui Fan,et al.  Principal modes in multimode waveguides. , 2005, Optics letters.

[17]  Sanjaya Lohani,et al.  On the use of deep neural networks in optical communications. , 2018, Applied optics.

[18]  Ivan Vishniakou,et al.  Light scattering control in transmission and reflection with neural networks. , 2018, Optics express.

[19]  D. Gloge,et al.  Optical power flow in multimode fibers , 1972 .

[20]  Yibo Zhang,et al.  Deep Learning Microscopy , 2017, ArXiv.

[21]  Demetri Psaltis,et al.  Multimode optical fiber transmission with a deep learning network , 2018, Light: Science & Applications.

[22]  S. Popoff,et al.  Using a multimode fiber as a high resolution, low loss spectrometer , 2013, 2013 Conference on Lasers & Electro-Optics Europe & International Quantum Electronics Conference CLEO EUROPE/IQEC.

[23]  Tomáš Čižmár,et al.  Seeing through chaos in multimode fibres , 2015, Nature Photonics.

[24]  Vincent Couderc,et al.  Spatial beam self-cleaning in multimode fibres , 2016, Nature Photonics.

[25]  S. Popoff,et al.  Measuring the transmission matrix in optics: an approach to the study and control of light propagation in disordered media. , 2009, Physical review letters.

[26]  Jürgen Schmidhuber,et al.  Deep learning in neural networks: An overview , 2014, Neural Networks.

[27]  Matthew D. Zeiler.  ADADELTA: An Adaptive Learning Rate Method , 2012, ArXiv.

[28]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[29]  Navid Borhani,et al.  Learning to see through multimode fibers , 2018, Optica.

[30]  K. Wagner,et al.  Adaptive wavefront shaping for controlling nonlinear multimode interactions in optical fibres , 2018 .

[31]  Ioannis N. Papadopoulos,et al.  Focusing and scanning light through a multimode optical fiber using digital phase conjugation. , 2012, Optics express.

[32]  K. Dholakia,et al.  Exploiting multimode waveguides for pure fibre-based imaging , 2012, Nature Communications.

[33]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[34]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[35]  Shuai Li,et al.  Lensless computational imaging through deep learning , 2017, ArXiv.

[36]  I. White,et al.  An experimental and theoretical study of the offset launch technique for the enhancement of the bandwidth of multimode fiber links , 1998 .

[37]  Lei Su,et al.  Bayes' theorem-based binary algorithm for fast reference-less calibration of a multimode fiber. , 2018, Optics express.

[38]  Brandon Redding,et al.  Using a multimode fiber as a high-resolution, low-loss spectrometer , 2012 .

[39]  Logan G. Wright,et al.  Controllable spatiotemporal nonlinear effects in multimode fibres , 2015, Nature Photonics.

[40]  Nathalie Japkowicz,et al.  The class imbalance problem: A systematic study , 2002, Intell. Data Anal..

[41]  Demetri Psaltis,et al.  High-resolution, lensless endoscope based on digital scanning through a multimode optical fiber , 2013, Biomedical optics express.

[42]  Lei Su,et al.  Characterization of an imaging multimode optical fiber using a digital micro-mirror device based single-beam system. , 2018, Optics express.

[43]  Liang Deng,et al.  Light Propagation Prediction through Multimode Optical Fibers with a Deep Neural Network , 2018, 2018 IEEE 3rd Advanced Information Technology, Electronic and Automation Control Conference (IAEAC).

[44]  Gregory Cohen,et al.  EMNIST: an extension of MNIST to handwritten letters , 2017, CVPR 2017.

[45]  Tomáš Čižmár,et al.  Three-dimensional holographic optical manipulation through a high-numerical-aperture soft-glass multimode fibre , 2018 .