期刊
OPTICS EXPRESS
卷 30, 期 7, 页码 10443-10455出版社
Optica Publishing Group
DOI: 10.1364/OE.443932
关键词
-
类别
资金
- ARC Future Fellowship [FT200100154]
- Optofab node of the Australian National Fabrication Facility utilizing Commonwealth and South Australian State Government
- ARC Centre for Nanoscale BioPhotonics [CE14010003]
- Fundacao de Amparo a Pesquisa do Estado de Sao Paulo [2018/10409-7]
- Australian Research Council [FT200100154] Funding Source: Australian Research Council
In this paper, we demonstrate that deep learning improves the analysis of fiber specklegrams for sensing, and apply it to measurements of air temperature and water immersion length. We compare deep neural networks (DNNs) including a convolutional neural network and a multi-layer perceptron network with a traditional correlation technique using data from a multimode fiber exposed-core fiber. We also show the ability of the DNNs to be trained against random noise sources like specklegram translations.
Fiber specklegram sensors (FSSs) traditionally use statistical methods to analyze specklegrams obtained from fibers for sensing purposes, but can suffer from limitations such as vulnerability to noise and lack of dynamic range. In this paper we demonstrate that deep learning improves the analysis of specklegrams for sensing, which we show here for both air temperature and water immersion length measurements. 'Pwo deep neural networks (DNNs); a convolutional neural network and a multi-layer perceptron network, are used and compared to a traditional correlation technique on data obtained from a multimode fiber exposed-core fiber. The ability for the DNNs to be trained against a random noise source such as specklegram translations is also demonstrated. (C) 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据