Evaluating the generalization capability of MobileNet-based CNNs for temperature prediction in simulated specklegram fiber optic sensors with combined synthetic datasets and data augmentation

The development of machine learning algorithms applied to specklegram-based sensors has facilitated the development of novel approaches for measuring several physical variables; however, most of these methods evaluate a single Fiber Specklegram Sensor (FSS) on a limited dataset. This paper assesses...

Full description

Autores:: Huertas-Montes, Isaac
V´élez, Francisco
Aristizabal, Víctor
Trujillo, Carlos
Herrera-Ramírez, Jorge

Tipo de recurso:: Article of journal

Fecha de publicación:: 2025

Institución:: Universidad Tecnológica de Bolívar

Repositorio:: Repositorio Institucional UTB

Idioma:: eng

id	UTB2_37989818194a5181616af7ffe673a9e6
oai_identifier_str	oai:repositorio.utb.edu.co:20.500.12585/14264
network_acronym_str	UTB2
network_name_str	Repositorio Institucional UTB
repository_id_str
dc.title.spa.fl_str_mv	Evaluating the generalization capability of MobileNet-based CNNs for temperature prediction in simulated specklegram fiber optic sensors with combined synthetic datasets and data augmentation
dc.title.translated.spa.fl_str_mv	Evaluating the generalization capability of MobileNet-based CNNs for temperature prediction in simulated specklegram fiber optic sensors with combined synthetic datasets and data augmentation
title	Evaluating the generalization capability of MobileNet-based CNNs for temperature prediction in simulated specklegram fiber optic sensors with combined synthetic datasets and data augmentation
spellingShingle	Evaluating the generalization capability of MobileNet-based CNNs for temperature prediction in simulated specklegram fiber optic sensors with combined synthetic datasets and data augmentation Specklegrams Convolutional Neural Network Finite element method FSS MobileNet
title_short	Evaluating the generalization capability of MobileNet-based CNNs for temperature prediction in simulated specklegram fiber optic sensors with combined synthetic datasets and data augmentation
title_full	Evaluating the generalization capability of MobileNet-based CNNs for temperature prediction in simulated specklegram fiber optic sensors with combined synthetic datasets and data augmentation
title_fullStr	Evaluating the generalization capability of MobileNet-based CNNs for temperature prediction in simulated specklegram fiber optic sensors with combined synthetic datasets and data augmentation
title_full_unstemmed	Evaluating the generalization capability of MobileNet-based CNNs for temperature prediction in simulated specklegram fiber optic sensors with combined synthetic datasets and data augmentation
title_sort	Evaluating the generalization capability of MobileNet-based CNNs for temperature prediction in simulated specklegram fiber optic sensors with combined synthetic datasets and data augmentation
dc.creator.fl_str_mv	Huertas-Montes, Isaac V´élez, Francisco Aristizabal, Víctor Trujillo, Carlos Herrera-Ramírez, Jorge
dc.contributor.author.eng.fl_str_mv	Huertas-Montes, Isaac V´élez, Francisco Aristizabal, Víctor Trujillo, Carlos Herrera-Ramírez, Jorge
dc.subject.eng.fl_str_mv	Specklegrams Convolutional Neural Network Finite element method FSS MobileNet
topic	Specklegrams Convolutional Neural Network Finite element method FSS MobileNet
description	The development of machine learning algorithms applied to specklegram-based sensors has facilitated the development of novel approaches for measuring several physical variables; however, most of these methods evaluate a single Fiber Specklegram Sensor (FSS) on a limited dataset. This paper assesses the generalization capability of applying these algorithms, in particular, Convolutional Neural Networks (CNN), to the prediction of temperature in simulated FSSs with different characteristics and conditions. This is achieved through the use of multiple combined synthetic datasets and data augmentation. The application of the Finite Element Method (FEM) enables the generation of datasets within the COMSOL Multi-physics software. The datasets are simulated with varying optical parameters, representing different optical fibers. Following the simulation of the datasets and training of selected models by combining them, data augmentation tests are conducted as though they were real fiber optic disturbances. Ultimately, a model is generated incorporating all the combined datasets and data augmentation, demonstrating the capacity of the model for generalization. This showcases the versatility of the computational methodology for evaluating, designing, and adjusting sensors without the need for experimental data. Additionally, it illustrates that a relatively simple model can be adapted to a variety of sensing system scenarios and configurations.
publishDate	2025
dc.date.accessioned.none.fl_str_mv	2025-09-15 00:00:00
dc.date.available.none.fl_str_mv	2025-09-15 00:00:00
dc.date.issued.none.fl_str_mv	2025-09-15
dc.type.spa.fl_str_mv	Artículo de revista
dc.type.coar.fl_str_mv	http://purl.org/coar/resource_type/c_2df8fbb1
dc.type.driver.eng.fl_str_mv	info:eu-repo/semantics/article
dc.type.coar.eng.fl_str_mv	http://purl.org/coar/resource_type/c_6501
dc.type.local.eng.fl_str_mv	Journal article
dc.type.content.eng.fl_str_mv	Text
dc.type.version.eng.fl_str_mv	info:eu-repo/semantics/publishedVersion
dc.type.coarversion.eng.fl_str_mv	http://purl.org/coar/version/c_970fb48d4fbd8a85
format	http://purl.org/coar/resource_type/c_6501
status_str	publishedVersion
dc.identifier.url.none.fl_str_mv	https://doi.org/10.32397/tesea.vol6.n2.802
dc.identifier.doi.none.fl_str_mv	10.32397/tesea.vol6.n2.802
dc.identifier.eissn.none.fl_str_mv	2745-0120
url	https://doi.org/10.32397/tesea.vol6.n2.802
identifier_str_mv	10.32397/tesea.vol6.n2.802 2745-0120
dc.language.iso.eng.fl_str_mv	eng
language	eng
dc.relation.references.eng.fl_str_mv	Jianli Liu, Yuxin Ke, Dong Yang, Qiao Deng, Chuang Hei, Hu Han, Daicheng Peng, Fangqing Wen, Ankang Feng, and Xueran Zhao. Deep learning-based simultaneous temperature- and curvature-sensitive scatterplot recognition. Sensors, 24(13), 2024. [2] Mohammad Istiaque Reja, Darcy L. Smith, Linh Viet Nguyen, Heike Ebendorff-Heidepriem, and Stephen C. Warren-Smith. Multimode optical fiber specklegram pressure sensor using deep learning. IEEE Transactions on Instrumentation and Measurement, 73:1–10, 2024. [3] Ivan Chapalo, Andreas Stylianou, Patrice Mégret, and Antreas Theodosiou. Advances in optical fiber speckle sensing: A comprehensive review. Photonics, 11(4), 2024. [4] Shichao Yue, Huizhen Lu, Boyi Li, and Zifan Che. Feasibility of a specklegram-based quasi-distributed temperature sensor with principal component analysis and variational autoencoder. IEEE Sensors Journal, 24(14):22410–22418, 2024. [5] Fang Zhao, Weihao Lin, Penglai Guo, Jie Hu, Yuhui Liu, Shuaiqi Liu, Feihong Yu, Guomeng Zuo, Guoqing Wang, Huanhuan Liu, Jinna Chen, Yi Li, Perry Ping Shum, and Liyang Shao. Compact optical fiber sensor based on vernier effect with speckle patterns. Opt. Express, 31(22):36940–36951, Oct 2023. [6] Hanchao Sun, Jixuan Wu, Binbin Song, Binbin Song, Jifang Wang, and Xiao Liu. Speckle-decoded temperature-insensitive strain identification in a multimode optical fiber. Optics Letters, Vol. 49, Issue 21, pp. 6185-6188, 49:6185–6188, 11 2024. [7] V. H. Arístizabal, F. J. Vélez, E. Rueda, N. D. Gómez, and J. A. Gómez. Numerical modeling of fiber specklegram sensors by using finite element method (fem). Opt. Express, 24(24):27225–27238, Nov 2016. [8] Juan Arango, Victor Aristizabal, Francisco Vélez, Juan Carrasquilla, Jorge Gomez, Jairo Quijano, and Jorge Herrera-Ramirez. Synthetic dataset of speckle images for fiber optic temperature sensor. Data in Brief, 48:109134, 2023. [9] Madhu Veettikazhy, Anders Kragh Hansen, Dominik Marti, Stefan Mark Jensen, Anja Lykke Borre, Esben Ravn Andresen, Kishan Dholakia, and Peter Eskil Andersen. Bpm-matlab: an open-source optical propagation simulation tool in matlab. Opt. Express, 29(8):11819–11832, Apr 2021. [10] Brian Pamukti, Zi Wang, Muhammad Fajar Faliasthiunus Pradipta, Shien-Kuei Liaw, Chien-Hung Yeh, and Fu-Liang Yang. Deep learning and time series signal processing for bending detection in mining environment using optical fiber sensor. Optical Fiber Technology, 88:103819, 2024. [11] Juan Sanguino-Lemus, Gustavo Hernández-Martínez, and Carla Puerto-López. Monitoreo estructural basado en sistemas de sensores de fibra óptica. Revista de Ingenierías Interfaces, 3:73–97, 2020. [12] Guangde Li, Yan Liu, Lezhi Pang, Hui Yuan, and Muguang Wang. A novel structure with ultra short multimode fiber for f iber specklegram sensor and its application in multi-bending sensing. proof and concept. Journal of Lightwave Technology, pages 1–9, 2024. [13] Asif Newaz, Md Omar Faruque, Rabiul Al Mahmud, Rakibul Hasan Sagor, and Mohammed Zahed Mustafa Khan. Machine-learning-enabled multimode fiber specklegram sensors: A review. IEEE Sensors Journal, 23(18):20937–20950, 2023. [14] Wataru Matsuda, Yuji Yuhara, Kaisei Sato, and Shinya Sasaki. A study on prediction of friction characteristics from speckle patterns of friction surfaces using machine learning. Tribology Online, 19(4):334–344, 2024. [15] Yuhui Liu, Weihao Lin, Fang Zhao, Yibin Liu, Junhui Sun, Jie Hu, Jialong Li, Jinna Chen, Xuming Zhang, Mang I. Vai, Perry Ping Shum, and Liyang Shao. A multimode microfiber specklegram biosensor for measurement of ceacam5 through ai diagnosis. Biosensors, 14(1), 2024. [16] Alberto Rodríguez-Cuevas, Eusebio Real Pena, Luis Rodríguez-Cobo, Mauro Lomer, and José Miguel López-Higuera. Low-cost fiber specklegram sensor for noncontact continuous patient monitoring. Journal of Biomedical Optics, 22(3):037001, 2017. [17] Laura Susana Vargas-Valencia, Felipe B. A. Schneider, Arnaldo G. Leal-Junior, Pablo Caicedo-Rodríguez, Wilson A. Sierra-Arévalo, Luis E. Rodríguez-Cheu, Teodiano Bastos-Filho, and Anselmo Frizera-Neto. Sleeve for knee angle monitoring: An imu-pof sensor fusion system. IEEE Journal of Biomedical and Health Informatics, 25(2):465–474, 2021. [18] Eric Fujiwara, Murilo Ferreira Marques dos Santos, and Carlos Kenichi Suzuki. Optical fiber specklegram sensor analysis by speckle pattern division. Applied Optics, 56:1585, 2 2017. [19] Fedor Gubarev, Lin Li, Miron Klenovskii, and Anatoliy Glotov. Speckle pattern processing by digital image correlation. MATECWebof Conferences, 48:04003, 4 2016. [20] Yan Liu, Guangde Li, Qi Qin, Zhongwei Tan, Muguang Wang, and Fengping Yan. Bending recognition based on the analysis of fiber specklegrams using deep learning. Optics Laser Technology, 131:106424, 11 2020. [21] Francisco Velez Hoyos, Juan David Arango Moreno, Victor Aristizabal, Carlos Trujillo, and Jorge Herrera Ramirez. Comparative performance evaluation of classical methods and a deep learning approach for temperature prediction in fiber optic specklegram sensors. Computer Optics, 48:689–695, 09 2024. [22] Xinliang Gao, Jixuan Wu, Binbin Song, Haifeng Liu, Shaoxiang Duan, Zhuo Zhang, Xiao Liu, and Hanchao Sun. Deep learning for temperature sensing with microstructure fiber in noise perturbation environment. IEEE Photonics Technology Letters, 35:1247–1250, 12 2023. [23] Darcy L. Smith, Linh V. Nguyen, David J. Ottaway, Thiago D. Cabral, Thiago D. Cabral, Eric Fujiwara, Cristiano M. B. Cordeiro, Cristiano M. B. Cordeiro, Stephen C. Warren-Smith, Stephen C. Warren-Smith, and Stephen C. Warren-Smith. Machine learning for sensing with a multimode exposed core fiber specklegram sensor. Optics Express, Vol. 30, Issue 7, pp. 10443-10455, 30:10443–10455, 3 2022. [24] Guangde Li, Yan Liu, Qi Qin, Xiaoli Zou, Muguang Wang, and Fengping Yan. Deep learning based optical curvature sensor through specklegram detection of multimode fiber. Optics Laser Technology, 149:107873, 5 2022. [25] Fu Feng, Jiaan Gan, PengFei Chen, Wei Lin, GuangYong Chen, Changjun Min, Xiaocong Yuan, and Michael Somekh. Ai-assisted spectrometer based on multi-mode optical fiber speckle patterns. Optics Communications, 522:128675, 2022. [26] Juan David Arango Moreno, Yeraldin Velez, Victor Aristizabal, Francisco Velez, Gómez Alberto, Jairo Quijano, and Jorge Herrera Ramirez. Numerical study using finite element method for the thermal response of fiber specklegram sensors with changes in the length of the sensing zone. Computer Optics, 45:534–540, 07 2021. [27] Juan Arango, Victor Aristizabal, Francisco Vélez, Juan Carrasquilla, Jorge Gomez, Jairo Quijano, and Jorge Herrera-Ramirez. Synthetic dataset of speckle images for fiber optic temperature sensor. Data in Brief, 48:109134, 6 2023. [28] Francisco J. Vélez, Juan D. Arango, Víctor H. Aristizábal, Carlos Trujillo, and Jorge A. Herrera-Ramírez. Experimental dataset for fiber optic specklegram sensing under thermal conditions and use in a deep learning interrogation scheme. Data, 10:44, 3 2025. [29] Luis Castaño, Luis Gutierrez, Jairo Quijano, Jorge Herrera-Ramírez, Alejandro Hoyos, Francisco Vélez, Víctor Aristizabal, Luiz Silva-Nunez, and Jorge Gómez. Temperature measurement by means of fiber specklegram sensors (fss). Óptica Pura y Aplicada, 2018. [30] Walter Frei. How large of a model can you solve with comsol®? COMSOL Blog, 2022. [31] COMSOL. Howto estimate the number of degrees of freedom in a model. COMSOL Learning Center, 2023.
dc.relation.ispartofjournal.eng.fl_str_mv	Transactions on Energy Systems and Engineering Applications
dc.relation.citationvolume.eng.fl_str_mv	6
dc.relation.citationstartpage.none.fl_str_mv	1
dc.relation.citationendpage.none.fl_str_mv	12
dc.relation.bitstream.none.fl_str_mv	https://revistas.utb.edu.co/tesea/article/download/802/459
dc.relation.citationedition.eng.fl_str_mv	Núm. 2 , Año 2025 : (In progress) Transactions on Energy Systems and Engineering Applications
dc.relation.citationissue.eng.fl_str_mv	2
dc.rights.uri.eng.fl_str_mv	https://creativecommons.org/licenses/by/4.0
dc.rights.accessrights.eng.fl_str_mv	info:eu-repo/semantics/openAccess
dc.rights.creativecommons.eng.fl_str_mv	This work is licensed under a Creative Commons Attribution 4.0 International License.
dc.rights.coar.eng.fl_str_mv	http://purl.org/coar/access_right/c_abf2
rights_invalid_str_mv	https://creativecommons.org/licenses/by/4.0 This work is licensed under a Creative Commons Attribution 4.0 International License. http://purl.org/coar/access_right/c_abf2
eu_rights_str_mv	openAccess
dc.format.mimetype.eng.fl_str_mv	application/pdf
dc.publisher.eng.fl_str_mv	Universidad Tecnológica de Bolívar
dc.source.eng.fl_str_mv	https://revistas.utb.edu.co/tesea/article/view/802
institution	Universidad Tecnológica de Bolívar
repository.name.fl_str_mv	Repositorio Digital Universidad Tecnológica de Bolívar
repository.mail.fl_str_mv	bdigital@metabiblioteca.com
_version_	1858228408199151616
spelling	Huertas-Montes, IsaacV´élez, FranciscoAristizabal, VíctorTrujillo, CarlosHerrera-Ramírez, Jorge2025-09-15 00:00:002025-09-15 00:00:002025-09-15The development of machine learning algorithms applied to specklegram-based sensors has facilitated the development of novel approaches for measuring several physical variables; however, most of these methods evaluate a single Fiber Specklegram Sensor (FSS) on a limited dataset. This paper assesses the generalization capability of applying these algorithms, in particular, Convolutional Neural Networks (CNN), to the prediction of temperature in simulated FSSs with different characteristics and conditions. This is achieved through the use of multiple combined synthetic datasets and data augmentation. The application of the Finite Element Method (FEM) enables the generation of datasets within the COMSOL Multi-physics software. The datasets are simulated with varying optical parameters, representing different optical fibers. Following the simulation of the datasets and training of selected models by combining them, data augmentation tests are conducted as though they were real fiber optic disturbances. Ultimately, a model is generated incorporating all the combined datasets and data augmentation, demonstrating the capacity of the model for generalization. This showcases the versatility of the computational methodology for evaluating, designing, and adjusting sensors without the need for experimental data. Additionally, it illustrates that a relatively simple model can be adapted to a variety of sensing system scenarios and configurations.application/pdfengUniversidad Tecnológica de BolívarIsaac Huertas-Montes, Francisco V´élez, Víctor Aristizabal, Carlos Trujillo, Jorge Herrera-Ramírez - 2025https://creativecommons.org/licenses/by/4.0info:eu-repo/semantics/openAccessThis work is licensed under a Creative Commons Attribution 4.0 International License.http://purl.org/coar/access_right/c_abf2https://revistas.utb.edu.co/tesea/article/view/802SpecklegramsConvolutional Neural NetworkFinite element methodFSSMobileNetEvaluating the generalization capability of MobileNet-based CNNs for temperature prediction in simulated specklegram fiber optic sensors with combined synthetic datasets and data augmentationEvaluating the generalization capability of MobileNet-based CNNs for temperature prediction in simulated specklegram fiber optic sensors with combined synthetic datasets and data augmentationArtículo de revistainfo:eu-repo/semantics/articlehttp://purl.org/coar/resource_type/c_6501http://purl.org/coar/resource_type/c_2df8fbb1Journal articleTextinfo:eu-repo/semantics/publishedVersionhttp://purl.org/coar/version/c_970fb48d4fbd8a85https://doi.org/10.32397/tesea.vol6.n2.80210.32397/tesea.vol6.n2.8022745-0120Jianli Liu, Yuxin Ke, Dong Yang, Qiao Deng, Chuang Hei, Hu Han, Daicheng Peng, Fangqing Wen, Ankang Feng, and Xueran Zhao. Deep learning-based simultaneous temperature- and curvature-sensitive scatterplot recognition. Sensors, 24(13), 2024. [2] Mohammad Istiaque Reja, Darcy L. Smith, Linh Viet Nguyen, Heike Ebendorff-Heidepriem, and Stephen C. Warren-Smith. Multimode optical fiber specklegram pressure sensor using deep learning. IEEE Transactions on Instrumentation and Measurement, 73:1–10, 2024. [3] Ivan Chapalo, Andreas Stylianou, Patrice Mégret, and Antreas Theodosiou. Advances in optical fiber speckle sensing: A comprehensive review. Photonics, 11(4), 2024. [4] Shichao Yue, Huizhen Lu, Boyi Li, and Zifan Che. Feasibility of a specklegram-based quasi-distributed temperature sensor with principal component analysis and variational autoencoder. IEEE Sensors Journal, 24(14):22410–22418, 2024. [5] Fang Zhao, Weihao Lin, Penglai Guo, Jie Hu, Yuhui Liu, Shuaiqi Liu, Feihong Yu, Guomeng Zuo, Guoqing Wang, Huanhuan Liu, Jinna Chen, Yi Li, Perry Ping Shum, and Liyang Shao. Compact optical fiber sensor based on vernier effect with speckle patterns. Opt. Express, 31(22):36940–36951, Oct 2023. [6] Hanchao Sun, Jixuan Wu, Binbin Song, Binbin Song, Jifang Wang, and Xiao Liu. Speckle-decoded temperature-insensitive strain identification in a multimode optical fiber. Optics Letters, Vol. 49, Issue 21, pp. 6185-6188, 49:6185–6188, 11 2024. [7] V. H. Arístizabal, F. J. Vélez, E. Rueda, N. D. Gómez, and J. A. Gómez. Numerical modeling of fiber specklegram sensors by using finite element method (fem). Opt. Express, 24(24):27225–27238, Nov 2016. [8] Juan Arango, Victor Aristizabal, Francisco Vélez, Juan Carrasquilla, Jorge Gomez, Jairo Quijano, and Jorge Herrera-Ramirez. Synthetic dataset of speckle images for fiber optic temperature sensor. Data in Brief, 48:109134, 2023. [9] Madhu Veettikazhy, Anders Kragh Hansen, Dominik Marti, Stefan Mark Jensen, Anja Lykke Borre, Esben Ravn Andresen, Kishan Dholakia, and Peter Eskil Andersen. Bpm-matlab: an open-source optical propagation simulation tool in matlab. Opt. Express, 29(8):11819–11832, Apr 2021. [10] Brian Pamukti, Zi Wang, Muhammad Fajar Faliasthiunus Pradipta, Shien-Kuei Liaw, Chien-Hung Yeh, and Fu-Liang Yang. Deep learning and time series signal processing for bending detection in mining environment using optical fiber sensor. Optical Fiber Technology, 88:103819, 2024. [11] Juan Sanguino-Lemus, Gustavo Hernández-Martínez, and Carla Puerto-López. Monitoreo estructural basado en sistemas de sensores de fibra óptica. Revista de Ingenierías Interfaces, 3:73–97, 2020. [12] Guangde Li, Yan Liu, Lezhi Pang, Hui Yuan, and Muguang Wang. A novel structure with ultra short multimode fiber for f iber specklegram sensor and its application in multi-bending sensing. proof and concept. Journal of Lightwave Technology, pages 1–9, 2024. [13] Asif Newaz, Md Omar Faruque, Rabiul Al Mahmud, Rakibul Hasan Sagor, and Mohammed Zahed Mustafa Khan. Machine-learning-enabled multimode fiber specklegram sensors: A review. IEEE Sensors Journal, 23(18):20937–20950, 2023. [14] Wataru Matsuda, Yuji Yuhara, Kaisei Sato, and Shinya Sasaki. A study on prediction of friction characteristics from speckle patterns of friction surfaces using machine learning. Tribology Online, 19(4):334–344, 2024. [15] Yuhui Liu, Weihao Lin, Fang Zhao, Yibin Liu, Junhui Sun, Jie Hu, Jialong Li, Jinna Chen, Xuming Zhang, Mang I. Vai, Perry Ping Shum, and Liyang Shao. A multimode microfiber specklegram biosensor for measurement of ceacam5 through ai diagnosis. Biosensors, 14(1), 2024. [16] Alberto Rodríguez-Cuevas, Eusebio Real Pena, Luis Rodríguez-Cobo, Mauro Lomer, and José Miguel López-Higuera. Low-cost fiber specklegram sensor for noncontact continuous patient monitoring. Journal of Biomedical Optics, 22(3):037001, 2017. [17] Laura Susana Vargas-Valencia, Felipe B. A. Schneider, Arnaldo G. Leal-Junior, Pablo Caicedo-Rodríguez, Wilson A. Sierra-Arévalo, Luis E. Rodríguez-Cheu, Teodiano Bastos-Filho, and Anselmo Frizera-Neto. Sleeve for knee angle monitoring: An imu-pof sensor fusion system. IEEE Journal of Biomedical and Health Informatics, 25(2):465–474, 2021. [18] Eric Fujiwara, Murilo Ferreira Marques dos Santos, and Carlos Kenichi Suzuki. Optical fiber specklegram sensor analysis by speckle pattern division. Applied Optics, 56:1585, 2 2017. [19] Fedor Gubarev, Lin Li, Miron Klenovskii, and Anatoliy Glotov. Speckle pattern processing by digital image correlation. MATECWebof Conferences, 48:04003, 4 2016. [20] Yan Liu, Guangde Li, Qi Qin, Zhongwei Tan, Muguang Wang, and Fengping Yan. Bending recognition based on the analysis of fiber specklegrams using deep learning. Optics Laser Technology, 131:106424, 11 2020. [21] Francisco Velez Hoyos, Juan David Arango Moreno, Victor Aristizabal, Carlos Trujillo, and Jorge Herrera Ramirez. Comparative performance evaluation of classical methods and a deep learning approach for temperature prediction in fiber optic specklegram sensors. Computer Optics, 48:689–695, 09 2024. [22] Xinliang Gao, Jixuan Wu, Binbin Song, Haifeng Liu, Shaoxiang Duan, Zhuo Zhang, Xiao Liu, and Hanchao Sun. Deep learning for temperature sensing with microstructure fiber in noise perturbation environment. IEEE Photonics Technology Letters, 35:1247–1250, 12 2023. [23] Darcy L. Smith, Linh V. Nguyen, David J. Ottaway, Thiago D. Cabral, Thiago D. Cabral, Eric Fujiwara, Cristiano M. B. Cordeiro, Cristiano M. B. Cordeiro, Stephen C. Warren-Smith, Stephen C. Warren-Smith, and Stephen C. Warren-Smith. Machine learning for sensing with a multimode exposed core fiber specklegram sensor. Optics Express, Vol. 30, Issue 7, pp. 10443-10455, 30:10443–10455, 3 2022. [24] Guangde Li, Yan Liu, Qi Qin, Xiaoli Zou, Muguang Wang, and Fengping Yan. Deep learning based optical curvature sensor through specklegram detection of multimode fiber. Optics Laser Technology, 149:107873, 5 2022. [25] Fu Feng, Jiaan Gan, PengFei Chen, Wei Lin, GuangYong Chen, Changjun Min, Xiaocong Yuan, and Michael Somekh. Ai-assisted spectrometer based on multi-mode optical fiber speckle patterns. Optics Communications, 522:128675, 2022. [26] Juan David Arango Moreno, Yeraldin Velez, Victor Aristizabal, Francisco Velez, Gómez Alberto, Jairo Quijano, and Jorge Herrera Ramirez. Numerical study using finite element method for the thermal response of fiber specklegram sensors with changes in the length of the sensing zone. Computer Optics, 45:534–540, 07 2021. [27] Juan Arango, Victor Aristizabal, Francisco Vélez, Juan Carrasquilla, Jorge Gomez, Jairo Quijano, and Jorge Herrera-Ramirez. Synthetic dataset of speckle images for fiber optic temperature sensor. Data in Brief, 48:109134, 6 2023. [28] Francisco J. Vélez, Juan D. Arango, Víctor H. Aristizábal, Carlos Trujillo, and Jorge A. Herrera-Ramírez. Experimental dataset for fiber optic specklegram sensing under thermal conditions and use in a deep learning interrogation scheme. Data, 10:44, 3 2025. [29] Luis Castaño, Luis Gutierrez, Jairo Quijano, Jorge Herrera-Ramírez, Alejandro Hoyos, Francisco Vélez, Víctor Aristizabal, Luiz Silva-Nunez, and Jorge Gómez. Temperature measurement by means of fiber specklegram sensors (fss). Óptica Pura y Aplicada, 2018. [30] Walter Frei. How large of a model can you solve with comsol®? COMSOL Blog, 2022. [31] COMSOL. Howto estimate the number of degrees of freedom in a model. COMSOL Learning Center, 2023.Transactions on Energy Systems and Engineering Applications6112https://revistas.utb.edu.co/tesea/article/download/802/459Núm. 2 , Año 2025 : (In progress) Transactions on Energy Systems and Engineering Applications220.500.12585/14264oai:repositorio.utb.edu.co:20.500.12585/142642025-11-06 09:15:13.246https://creativecommons.org/licenses/by/4.0Isaac Huertas-Montes, Francisco V´élez, Víctor Aristizabal, Carlos Trujillo, Jorge Herrera-Ramírez - 2025metadata.onlyhttps://repositorio.utb.edu.coRepositorio Digital Universidad Tecnológica de Bolívarbdigital@metabiblioteca.com

Evaluating the generalization capability of MobileNet-based CNNs for temperature prediction in simulated specklegram fiber optic sensors with combined synthetic datasets and data augmentation

Publicaciones similares