Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/135076
Citations | ||
Scopus | Web of Science® | Altmetric |
---|---|---|
?
|
?
|
Type: | Journal article |
Title: | Machine learning for sensing with a multimode exposed core fiber specklegram sensor |
Author: | SMITH, D.L. NGUYEN, L.V. OTTAWAY, D.J. CABRAL, T.D. Fujiwara, E. CORDEIRO, C.M.B. WARREN-SMITH, S.C. |
Citation: | Optics Express, 2022; 30(7):1-13 |
Publisher: | Optica Publishing Group |
Issue Date: | 2022 |
ISSN: | 1094-4087 1094-4087 |
Statement of Responsibility: | Darcy L. Smith, Linh V. Nguyen, David J Ottoway, Thiago D. Cabral, Eric Fujiwara, Cristiano M.B. Cordeiro, Stephen C. Warren-Smith |
Abstract: | 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. Two 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. |
Rights: | © 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement |
DOI: | 10.1364/OE.443932 |
Grant ID: | http://purl.org/au-research/grants/arc/FT200100154 http://purl.org/au-research/grants/arc/CE14010003 |
Published version: | http://dx.doi.org/10.1364/oe.443932 |
Appears in Collections: | IPAS publications Physics publications |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
hdl_135076.pdf | Published version | 3.26 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.