Deep learning assisted XRF spectra classification

• Published in: Scientific reports 2024-02, Vol.14 (1), p.3666-10, Article 3666
• Main Authors: Andric, Velibor, Kvascev, Goran, Cvetanovic, Milos, Stojanovic, Sasa, Bacanin, Nebojsa, Gajic-Kvascev, Maja
• Format: Article
• Language: English
• Subjects: 639/166/987; 639/301/930/12; AI classification; Artificial intelligence; Autoencoder neural network; Chemical composition; Classification; Cultural heritage; Deep learning; Dimension reduction; Humanities and Social Sciences; multidisciplinary; Multivariate analysis; Neural networks; Pigments; Science; Science (multidisciplinary); Spectrometry; XRF spectra
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-024-53988-z

EDXRF spectrometry is a well-established and often-used analytical technique in examining materials from which cultural heritage objects are made. The analytical results are traditionally subjected to additional multivariate analysis for archaeometry studies to reduce the initial data’s dimensionality based on informative features. Nowadays, artificial intelligence (AI) techniques are used more for this purpose. Different soft computing techniques are used to improve speed and accuracy. Choosing the most suitable AI method can increase the sustainability of the analytical process and postprocessing activities. An autoencoder neural network has been designed and used as a dimension reduction tool of initial 40 × 2048 data collected in the raw EDXRF spectra, containing information about the selected points’ elemental composition on the canvas paintings’ surface. The autoencoder network design enables the best possible reconstruction of the original EDXRF spectrum and the most informative feature extraction, which has been used for dimension reduction. Such configuration allows for efficient classification algorithms and their performances. The autoencoder neural network approach is more sustainable, especially in processing time consumption and experts’ manual work.