Detail publikace

Spectral library transfer between distinct Laser-Induced Breakdown Spectroscopy systems trained on simultaneous measurements

VRÁBEL, J. KÉPEŠ, E. NEDĚLNÍK, P. BUDAY, J. CEMPÍREK, J. POŘÍZKA, P. KAISER, J.

Originální název

Spectral library transfer between distinct Laser-Induced Breakdown Spectroscopy systems trained on simultaneous measurements

Typ

článek v časopise ve Web of Science, Jimp

Jazyk

angličtina

Originální abstrakt

The mutual incompatibility of distinct spectroscopic systems is among the most limiting factors in Laser-Induced Breakdown Spectroscopy (LIBS). The cost related to setting up a new LIBS system is increased, as its extensive calibration is required. Solving the problem would enable inter-laboratory reference measurements and shared spectral libraries, which are fundamental for other spectroscopic techniques. We study a simplified version of this challenge where LIBS systems differ only in used spectrometers and collection optics but share all other parts of the apparatus, and collect spectra simultaneously from the same plasma plume. Extensive datasets measured as hyperspectral images of heterogeneous rock sample are used to train machine learning models that can transfer spectra between systems. The transfer is realized by a composed model that consists of a variational autoencoder (VAE) and a multilayer perceptron (MLP). The VAE is used to create a latent representation of spectra from the Primary system. Subsequently, spectra from the Secondary system are mapped to corresponding locations in the latent space by the MLP. The transfer is evaluated by several figures of merit (Euclidean and cosine distances, both spatially resolved; k-means clustering of transferred spectra). We demonstrate the viability of the method and compare it to several baseline approaches of varying complexity.

Klíčová slova

spectroscopic data, library transfer, machine learning, artificial neural networks, autoencoder

Autoři

VRÁBEL, J.; KÉPEŠ, E.; NEDĚLNÍK, P.; BUDAY, J.; CEMPÍREK, J.; POŘÍZKA, P.; KAISER, J.

Vydáno

5. 4. 2023

Nakladatel

Royal Society of Chemistry

ISSN

1364-5544

Periodikum

Journal of Analytical Atomic Spectrometry

Ročník

38

Číslo

4

Stát

Spojené království Velké Británie a Severního Irska

Strany od

841

Strany do

853

Strany počet

12

URL

Plný text v Digitální knihovně

BibTex

@article{BUT182920,
  author="Jakub {Vrábel} and Erik {Képeš} and Pavel {Nedělník} and Jakub {Buday} and Jan {Cempírek} and Pavel {Pořízka} and Jozef {Kaiser}",
  title="Spectral library transfer between distinct Laser-Induced Breakdown Spectroscopy systems trained on simultaneous measurements",
  journal="Journal of Analytical Atomic Spectrometry",
  year="2023",
  volume="38",
  number="4",
  pages="841--853",
  doi="10.1039/D2JA00406B",
  issn="1364-5544",
  url="https://pubs.rsc.org/en/content/articlelanding/2023/ja/d2ja00406b"
}