All-fiber high-speed image detection enabled by deep learning

Liu, Zhoutian; Wang, Lele; Meng, Yuan; He, Tiantian; He, Sifeng; Yang, Yousi; Wang, Liuyue; Tian, Jiemo; Li, Dan; Yan, Ping; Gong, Mali; Liu, Qiang; Xiao, Qirong

doi:10.1038/s41467-022-29178-8

Cited by 50 publications

(25 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Multimode nonlinear optical (MMNLO) systems have served as testbeds for exploring numerous physical phenomena to inspire communication, [1][2][3] imaging, [4][5][6] computing, [7][8][9] and lasers. [10][11][12] Multimode fiber (MMF) as an ideal MMNLO system has become a hotspot in recent years because of its flexible structure, plenty of modal channels, and abundant nonlinear DOI: 10.1002/lpor.202200987 interactions.…”

Section: Introductionmentioning

confidence: 99%

Beam Self‐Cleaning of 1.5 μ$\umu$m High Peak‐Power Spatiotemporal Mode‐Locked Lasers Enabled By Nonlinear Compression and Disorder

et al. 2023

Laser & Photonics Reviews

Self Cite

View full text Add to dashboard Cite

The realization of beam self-cleaning in a cavity is more challenging than in the outside cavity. The peak power of intracavity pulses needs to be high to reach the threshold of beam self-cleaning, which usually relies on additional diffraction grating to compress pulses in a positive-dispersion cavity. Here, it is first experimentally and numerically demonstrated that self-cleaning can be observed in an all-fiber high peak-power Er-doped spatiotemporal mode-locked (STML) laser at all-negative-dispersion. Through the nonlinear compression of graded-index multimode fiber, the pulses are compressed along with the emergence of beam self-cleaning. Besides, the inherent disorder of multimode fiber accelerates the self-cleaning process. The intracavity pulse energy of ≈13 nJ with a pulse duration of 734.5 fs is derived under a highly multimode excitation, with an output pulse energy of 2.33 nJ. The pulse energy is a nearly fourfold improvement over the previous report in all-fiber STML at 1.5 𝛍m. Temporal-dependent characteristics and nonlinear polarization dynamics of beam self-cleaning are also experimentally uncovered. It is demonstrated that the STML fiber laser will enable new insights into nonlinear pulse propagation in cavities and related applications.

show abstract

Section: Introductionmentioning

confidence: 99%

Beam Self‐Cleaning of 1.5 μ$\umu$m High Peak‐Power Spatiotemporal Mode‐Locked Lasers Enabled By Nonlinear Compression and Disorder

et al. 2023

Laser & Photonics Reviews

Self Cite

View full text Add to dashboard Cite

show abstract

“…Recently, deep learning approaches [22][23][24][25][26][27][28][29][30][31][32] have emerged that infer images from speckle patterns without prior knowledge of the fiber characteristics. Although they demonstrate promising results and real-time reconstruction, accounting for fiber bending remains a challenging problem.…”

Section: Introductionmentioning

confidence: 99%

Robust real-time imaging through flexible multimode fibers

Abdulaziz¹,

Mekhail

Altmann³

et al. 2023

Preprint

View full text Add to dashboard Cite

Conventional endoscopes comprise a bundle of optical fibers, associating one fiber for each pixel in the image. In principle, this can be reduced to a single multimode optical fiber (MMF), the width of a human hair, with one fiber spatial-mode per image pixel. However, images transmitted through a MMF emerge as unrecognisable speckle patterns due to dispersion and coupling between the spatial modes of the fiber. Furthermore, speckle patterns change as the fiber undergoes bending, making the use of MMFs in flexible imaging applications even more complicated. In this paper, we propose a real-time imaging system using flexible MMFs, but which is robust to bending. Our approach does not require access or feedback signal from the distal end of the fiber during imaging. We leverage a variational autoencoder (VAE) to reconstruct and classify images from the speckles and show that these images can still be recovered when the bend configuration of the fiber is changed to one that was not part of the training set. We utilize a MMF 300 mm long with a 50 um core for imaging 10x10 cm objects placed approximately at 20 cm from the fiber and the system can deal with a change in fiber bend of 50 degrees and range of movement of 8 cm.

show abstract

“…[31,32] Recently, time-of-flight based 3D imaging and mode dispersion based high-speed detection have also been stunningly demonstrated. [33,34] MMFs convert image information into combinations of transverse modes, making the information capacity an order of magnitude larger than the number of cores in CFBs with the same diameter. [35] Therefore, MMF imaging technology largely solves the dilemma of incompatibility between small fiber size and high imaging resolution encountered by imaging systems with CFBs.…”

Section: Introductionmentioning

confidence: 99%

High‐Speed All‐Fiber Micro‐Imaging with Large Depth of Field

Wang

Yang

Liu

et al. 2022

Laser & Photonics Reviews

Self Cite

View full text Add to dashboard Cite

Single fiber imaging has evolved into a powerful method for detecting minute objects in narrow spaces. However, existing systems are not conducive to imaging dynamic objects at depth due to their bulky probes, time-consuming scanning acquisition methods, and transmissive illumination mode. Minimally invasive reflection mode imaging with high spatial and temporal resolution remains an open challenge. Here, a precise and high-speed imaging scheme without scanning is proposed. Multimode fiber imaging technology is incorporated into an all-fiber aberration-free precision detection system. High temporal resolution (5000 fps) detection of tiny natural scenes is experimentally realized by optimizing the approximation of the inverse transmission matrix in a simple and compact setup. The system can display the detected screen in real-time and the computational imaging with a large depth of field (1 mm) is enabled by jointly learning. The recovery results are superior compared to typical deep neural networks. The demonstrated scheme offers a new possibility for many applications, for example, microendoscopy, all-optical computing, and remote high-speed video transmission.

show abstract

All-fiber high-speed image detection enabled by deep learning

Cited by 50 publications

References 41 publications

Beam Self‐Cleaning of 1.5 μ$\umu$m High Peak‐Power Spatiotemporal Mode‐Locked Lasers Enabled By Nonlinear Compression and Disorder

Beam Self‐Cleaning of 1.5 μ$\umu$m High Peak‐Power Spatiotemporal Mode‐Locked Lasers Enabled By Nonlinear Compression and Disorder

Robust real-time imaging through flexible multimode fibers

High‐Speed All‐Fiber Micro‐Imaging with Large Depth of Field

Contact Info

Product

Resources

About