Abstract

Due to the occurrence of redundant speckle, multimode fiber (MMF) imaging is extremely challenging. Our work studies the relationship between the effective feature distribution of the speckle field and the local spatial position and area, and proves that the information distribution of the speckle is highly redundant. The effective feature refers to the phase and amplitude information of the optical field carrying the image point information and the co-exciting very redundant information due to mode dispersion, interference, coupling, and entrained noise through transmission. The neural network Swin-Unet can well learn the association information between global and local features, greatly simplifies the fitting of the MMF end-to-end global mapping relationship, and achieves high-fidelity reconstruction from the local speckle field to the global image. This work will contribute to the realization of MMF real-time large-field endoscopic imaging.

© 2023 Optica Publishing Group

Deep learning image transmission through a multimode fiber based on a small training dataset

Binbin Song, Chang Jin, Jixuan Wu, Wei Lin, Bo Liu, Wei Huang, and Shengyong Chen
Opt. Express 30(4) 5657-5672 (2022)

Binary amplitude-only image reconstruction through a MMF based on an AE-SNN combined deep learning model

Hui Chen, Zhengquan He, Zaikun Zhang, Yi Geng, and Weixing Yu
Opt. Express 28(20) 30048-30062 (2020)

Self-supervised Blind2Unblind deep learning scheme for OCT speckle reductions

Xiaojun Yu, Chenkun Ge, Mingshuai Li, Miao Yuan, Linbo Liu, Jianhua Mo, Perry Ping Shum, and Jinna Chen
Biomed. Opt. Express 14(6) 2773-2795 (2023)

Simultaneous illumination and imaging based on a single multimode fiber

Zhenyu Ju, Zhenming Yu, Ziyi Meng, Ning Zhan, Lili Gui, and Kun Xu
Opt. Express 30(9) 15596-15606 (2022)

Deep learning-based object classification through multimode fiber via a CNN-architecture SpeckleNet

Ping Wang and Jianglei Di
Appl. Opt. 57(28) 8258-8263 (2018)