Self-Denoising of BOTDA Using Deep Convolutional Neural Networks

We propose the self-denoising network (SDNet), a self-supervised network based on a convolutional neural network (CNN), for Brillouin trace denoising. With the target noisy image as the only input, the proposed method has no hardware restriction, requirement for image priors, or assumption for noise...

Full description

Saved in:
Bibliographic Details
Main Authors: Di Qi, Chun-Kit Chan, Xun Guan
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Photonics Journal
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10972307/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850281654365454336
author Di Qi
Chun-Kit Chan
Xun Guan
author_facet Di Qi
Chun-Kit Chan
Xun Guan
author_sort Di Qi
collection DOAJ
description We propose the self-denoising network (SDNet), a self-supervised network based on a convolutional neural network (CNN), for Brillouin trace denoising. With the target noisy image as the only input, the proposed method has no hardware restriction, requirement for image priors, or assumption for noise distribution. The Bernoulli mask and the partial convolutional layer implemented in the encoding process help capture the input features efficiently, while the dropout in the decoding process extends the feasibility and generality of the proposed network. Experimental results indicate that for Brillouin traces with frequency steps of 0.5 MHz/1 MHz, the proposed SDNet can improve the accuracy of Brillouin frequency shift (BFS) estimation by 40%/61%, 32%/31% and 24%/32% under input signal-to-noise ratios (SNR) of 5.5 dB, 8.5 dB, and 11.5 dB, respectively, without degradation in spatial resolution. Furthermore, the SDNet demonstrates a robust denoising performance in BOTDA systems with different application scenarios.
format Article
id doaj-art-c75501d9004d4b1ea93d759e357f64f7
institution OA Journals
issn 1943-0655
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Photonics Journal
spelling doaj-art-c75501d9004d4b1ea93d759e357f64f72025-08-20T01:48:14ZengIEEEIEEE Photonics Journal1943-06552025-01-0117311010.1109/JPHOT.2025.356340510972307Self-Denoising of BOTDA Using Deep Convolutional Neural NetworksDi Qi0Chun-Kit Chan1https://orcid.org/0000-0002-7046-5335Xun Guan2https://orcid.org/0000-0001-5045-6004Lightwave Communications Laboratory, Department of Information Engineering, The Chinese University of Hong Kong, Hong KongLightwave Communications Laboratory, Department of Information Engineering, The Chinese University of Hong Kong, Hong KongTsinghua Shenzhen International Graduate School, Tsinghua University, Shenzhen, ChinaWe propose the self-denoising network (SDNet), a self-supervised network based on a convolutional neural network (CNN), for Brillouin trace denoising. With the target noisy image as the only input, the proposed method has no hardware restriction, requirement for image priors, or assumption for noise distribution. The Bernoulli mask and the partial convolutional layer implemented in the encoding process help capture the input features efficiently, while the dropout in the decoding process extends the feasibility and generality of the proposed network. Experimental results indicate that for Brillouin traces with frequency steps of 0.5 MHz/1 MHz, the proposed SDNet can improve the accuracy of Brillouin frequency shift (BFS) estimation by 40%/61%, 32%/31% and 24%/32% under input signal-to-noise ratios (SNR) of 5.5 dB, 8.5 dB, and 11.5 dB, respectively, without degradation in spatial resolution. Furthermore, the SDNet demonstrates a robust denoising performance in BOTDA systems with different application scenarios.https://ieeexplore.ieee.org/document/10972307/BOTDAsingle-image denoisingself-supervised learningSNR enhancementpartial convolutional neural network (CNN)
spellingShingle Di Qi
Chun-Kit Chan
Xun Guan
Self-Denoising of BOTDA Using Deep Convolutional Neural Networks
IEEE Photonics Journal
BOTDA
single-image denoising
self-supervised learning
SNR enhancement
partial convolutional neural network (CNN)
title Self-Denoising of BOTDA Using Deep Convolutional Neural Networks
title_full Self-Denoising of BOTDA Using Deep Convolutional Neural Networks
title_fullStr Self-Denoising of BOTDA Using Deep Convolutional Neural Networks
title_full_unstemmed Self-Denoising of BOTDA Using Deep Convolutional Neural Networks
title_short Self-Denoising of BOTDA Using Deep Convolutional Neural Networks
title_sort self denoising of botda using deep convolutional neural networks
topic BOTDA
single-image denoising
self-supervised learning
SNR enhancement
partial convolutional neural network (CNN)
url https://ieeexplore.ieee.org/document/10972307/
work_keys_str_mv AT diqi selfdenoisingofbotdausingdeepconvolutionalneuralnetworks
AT chunkitchan selfdenoisingofbotdausingdeepconvolutionalneuralnetworks
AT xunguan selfdenoisingofbotdausingdeepconvolutionalneuralnetworks