Neural network-based classification of X-ray fluorescence spectra of artists’ pigments: an approach leveraging a synthetic dataset created using the fundamental parameters method

Abstract X-ray fluorescence (XRF) spectroscopy is an analytical technique used to identify chemical elements that has found widespread use in the cultural heritage sector to characterise artists' materials including the pigments in paintings. It generates a spectrum with characteristic emission...

Full description

Bibliographic Details
Main Authors: Cerys Jones, Nathan S. Daly, Catherine Higgitt, Miguel R. D. Rodrigues
Format: Article
Language:English
Published: SpringerOpen 2022-06-01
Series:Heritage Science
Subjects:
Online Access:https://doi.org/10.1186/s40494-022-00716-3
_version_ 1818228248829493248
author Cerys Jones
Nathan S. Daly
Catherine Higgitt
Miguel R. D. Rodrigues
author_facet Cerys Jones
Nathan S. Daly
Catherine Higgitt
Miguel R. D. Rodrigues
author_sort Cerys Jones
collection DOAJ
description Abstract X-ray fluorescence (XRF) spectroscopy is an analytical technique used to identify chemical elements that has found widespread use in the cultural heritage sector to characterise artists' materials including the pigments in paintings. It generates a spectrum with characteristic emission lines relating to the elements present, which is interpreted by an expert to understand the materials therein. Convolutional neural networks (CNNs) are an effective method for automating such classification tasks—an increasingly important feature as XRF datasets continue to grow in size—but they require large libraries that capture the natural variation of each class for training. As an alternative to having to acquire such a large library of XRF spectra of artists' materials a physical model, the Fundamental Parameters (FP) method, was used to generate a synthetic dataset of XRF spectra representative of pigments typically encountered in Renaissance paintings that could then be used to train a neural network. The synthetic spectra generated—modelled as single layers of individual pigments—had characteristic element lines closely matching those found in real XRF spectra. However, as the method did not incorporate effects from the X-ray source, the synthetic spectra lacked the continuum and Rayleigh and Compton scatter peaks. Nevertheless, the network trained on the synthetic dataset achieved 100% accuracy when tested on synthetic XRF data. Whilst this initial network only attained 55% accuracy when tested on real XRF spectra obtained from reference samples, applying transfer learning using a small quantity of such real XRF spectra increased the accuracy to 96%. Due to these promising results, the network was also tested on select data acquired during macro XRF (MA-XRF) scanning of a painting to challenge the model with noisier spectra Although only tested on spectra from relatively simple paint passages, the results obtained suggest that the FP method can be used to create accurate synthetic XRF spectra of individual artists' pigments, free from X-ray tube effects, on which a classification model could be trained for application to real XRF data and that the method has potential to be extended to deal with more complex paint mixtures and stratigraphies.
first_indexed 2024-12-12T09:59:42Z
format Article
id doaj.art-e80112b4d3c546a293f5e14ca99dad43
institution Directory Open Access Journal
issn 2050-7445
language English
last_indexed 2024-12-12T09:59:42Z
publishDate 2022-06-01
publisher SpringerOpen
record_format Article
series Heritage Science
spelling doaj.art-e80112b4d3c546a293f5e14ca99dad432022-12-22T00:28:01ZengSpringerOpenHeritage Science2050-74452022-06-0110111410.1186/s40494-022-00716-3Neural network-based classification of X-ray fluorescence spectra of artists’ pigments: an approach leveraging a synthetic dataset created using the fundamental parameters methodCerys Jones0Nathan S. Daly1Catherine Higgitt2Miguel R. D. Rodrigues3Department of Electronic and Electrical Engineering, University College LondonNational GalleryNational GalleryDepartment of Electronic and Electrical Engineering, University College LondonAbstract X-ray fluorescence (XRF) spectroscopy is an analytical technique used to identify chemical elements that has found widespread use in the cultural heritage sector to characterise artists' materials including the pigments in paintings. It generates a spectrum with characteristic emission lines relating to the elements present, which is interpreted by an expert to understand the materials therein. Convolutional neural networks (CNNs) are an effective method for automating such classification tasks—an increasingly important feature as XRF datasets continue to grow in size—but they require large libraries that capture the natural variation of each class for training. As an alternative to having to acquire such a large library of XRF spectra of artists' materials a physical model, the Fundamental Parameters (FP) method, was used to generate a synthetic dataset of XRF spectra representative of pigments typically encountered in Renaissance paintings that could then be used to train a neural network. The synthetic spectra generated—modelled as single layers of individual pigments—had characteristic element lines closely matching those found in real XRF spectra. However, as the method did not incorporate effects from the X-ray source, the synthetic spectra lacked the continuum and Rayleigh and Compton scatter peaks. Nevertheless, the network trained on the synthetic dataset achieved 100% accuracy when tested on synthetic XRF data. Whilst this initial network only attained 55% accuracy when tested on real XRF spectra obtained from reference samples, applying transfer learning using a small quantity of such real XRF spectra increased the accuracy to 96%. Due to these promising results, the network was also tested on select data acquired during macro XRF (MA-XRF) scanning of a painting to challenge the model with noisier spectra Although only tested on spectra from relatively simple paint passages, the results obtained suggest that the FP method can be used to create accurate synthetic XRF spectra of individual artists' pigments, free from X-ray tube effects, on which a classification model could be trained for application to real XRF data and that the method has potential to be extended to deal with more complex paint mixtures and stratigraphies.https://doi.org/10.1186/s40494-022-00716-3X-ray fluorescenceConvolutional neural networksDeep learningTransfer learningClassificationSynthetic data
spellingShingle Cerys Jones
Nathan S. Daly
Catherine Higgitt
Miguel R. D. Rodrigues
Neural network-based classification of X-ray fluorescence spectra of artists’ pigments: an approach leveraging a synthetic dataset created using the fundamental parameters method
Heritage Science
X-ray fluorescence
Convolutional neural networks
Deep learning
Transfer learning
Classification
Synthetic data
title Neural network-based classification of X-ray fluorescence spectra of artists’ pigments: an approach leveraging a synthetic dataset created using the fundamental parameters method
title_full Neural network-based classification of X-ray fluorescence spectra of artists’ pigments: an approach leveraging a synthetic dataset created using the fundamental parameters method
title_fullStr Neural network-based classification of X-ray fluorescence spectra of artists’ pigments: an approach leveraging a synthetic dataset created using the fundamental parameters method
title_full_unstemmed Neural network-based classification of X-ray fluorescence spectra of artists’ pigments: an approach leveraging a synthetic dataset created using the fundamental parameters method
title_short Neural network-based classification of X-ray fluorescence spectra of artists’ pigments: an approach leveraging a synthetic dataset created using the fundamental parameters method
title_sort neural network based classification of x ray fluorescence spectra of artists pigments an approach leveraging a synthetic dataset created using the fundamental parameters method
topic X-ray fluorescence
Convolutional neural networks
Deep learning
Transfer learning
Classification
Synthetic data
url https://doi.org/10.1186/s40494-022-00716-3
work_keys_str_mv AT cerysjones neuralnetworkbasedclassificationofxrayfluorescencespectraofartistspigmentsanapproachleveragingasyntheticdatasetcreatedusingthefundamentalparametersmethod
AT nathansdaly neuralnetworkbasedclassificationofxrayfluorescencespectraofartistspigmentsanapproachleveragingasyntheticdatasetcreatedusingthefundamentalparametersmethod
AT catherinehiggitt neuralnetworkbasedclassificationofxrayfluorescencespectraofartistspigmentsanapproachleveragingasyntheticdatasetcreatedusingthefundamentalparametersmethod
AT miguelrdrodrigues neuralnetworkbasedclassificationofxrayfluorescencespectraofartistspigmentsanapproachleveragingasyntheticdatasetcreatedusingthefundamentalparametersmethod