Fibre-optic based particle sensing via deep learning

James A Grant-Jacob,Michael D T Mcdonnell,Yunhui Xie,Robert W Eason,Ben Mills,Benita S Mackay,Matthew Loxham,Matthew Praeger,David J Richardson,Saurabh Jain

doi:10.1088/2515-7647/ab437b

Abstract

We demonstrate the capability for the identification of single particles, via a neural network, directly from the backscattered light collected by a 30-core optical fibre, when particles are illuminated using a single mode fibre-coupled laser light source. The neural network was shown to be able to determine the specific species of pollen with ∼97% accuracy, along with the distance between the end of the 30-core sensing fibre and the particles, with an associated error of ±6 μm. The ability to be able to classify particles directly from backscattered light using an optical fibre has potential in environments in which transmission imaging is neither possible nor suitable, such as sensing over opaque media, in the deep sea or outer space.

Highlights

Fibre-optic based sensors are ideal for worldwide deployment and sensing in a range of in-situ environments owing to their low-cost, light-weight, small and flexible nature
They can operate via single-ended interrogation, such that one end of the fibre can be free to interact like a probe, allowing for in-situ sensing in a variety of environments [5]
We show that the neural network can determine the distance between the pollen particles and the end of the fibres, and we examine the robustness of the network by varying the ambient light levels using an additional white light source

Summary

Introduction

Fibre-optic based sensors are ideal for worldwide deployment and sensing in a range of in-situ environments owing to their low-cost, light-weight, small and flexible nature. We extend on our previous work where deep learning pollution particle detection was carried out using free-space optics [23], and other work showing the capture and analysis of particles in the field using free-space optics [65], by demonstrating the ability to successfully classify real-world bio-aerosol particles (pollen grains) in real-time, via collection of their backscattered light using optical fibres.

Results

Conclusion