Skip to main navigation Skip to search Skip to main content

Sectorial-Perturbation analysis of fiber specklegram using machine learning techniques

  • M. Fontana
  • , A. Rodríguez-Cuevas
  • , L. Rodríguez-Cobo
  • , J. Mateo
  • , M. Lomer*
  • *Corresponding author for this work
  • Universidad de Cantabria
  • Biomateriales y Nanomedicina (CIBER-BBN)
  • Instituto de Investigación Sanitaria Valdecilla (IDIVAL)
  • Universidad de Zaragoza

Research output: Chapter in Book/Published conference outputConference publication

Abstract

Fiber-optic specklegram patterns are highly sensitive to external perturbations such as vibrations, temperature, or strain. However, due to its highly random behaviour using this kind of system for distributed sensing remains. In this work, a distributed sensing speckle system has been designed. This proof-of-concept has been developed by perturbating a multimode plastic optical fiber in three different places, recording the videos of these perturbations and using them for training and testing machine learning algorithms. The results show classifications over 99% of accuracy when testing new data under certain conditions.

Original languageEnglish
Title of host publication26th International Conference on Plastic Optical Fibres, POF 2017 - Proceedings
EditorsAna Maria Rocha, Rogerio Nunes Nogueira
ISBN (Electronic)9789899734524
Publication statusPublished - 2017
Event26th International Conference on Plastic Optical Fibres, POF 2017 - Proceedings - Aveiro, Portugal
Duration: 13 Sept 201715 Sept 2017

Publication series

Name26th International Conference on Plastic Optical Fibres, POF 2017 - Proceedings

Conference

Conference26th International Conference on Plastic Optical Fibres, POF 2017 - Proceedings
Country/TerritoryPortugal
CityAveiro
Period13/09/1715/09/17

Fingerprint

Dive into the research topics of 'Sectorial-Perturbation analysis of fiber specklegram using machine learning techniques'. Together they form a unique fingerprint.

Cite this