Classification of Mineral Phases in Unpolished Samples Using a Transformer Neural Network

Classification of Mineral Phases in Unpolished Samples Using a Transformer Neural Network

Stať ve sborníku v databázi WoS či Scopus

Scanning electron microscopy (SEM) combined with energy dispersive spectroscopy (EDS) is widely used in geosciences for mineral phase classification. However, the lack of large, labeled datasets—especially for unpolished samples common in forensic pedology—limits the direct application of traditional supervised machine learning methods. In this study, we investigate the use of synthetic EDS data, generated using DTSA-II, to train advanced neural network architectures and evaluate their performance on real SEM–EDS measurements. We compare a U-Net baseline with a proposed 3D ResNet and a Transformer model. All models were trained on synthetic data and tested on real measurements of ten selected mineral phases. U-Net achieved accuracy of 63.7%, 3D ResNet reached 92.3%, and the Transformer model achieved the highest accuracy of 97.0%. These findings demonstrate that Transformer architectures can effectively generalize from synthetic to real EDS data, offering a promising way for accurate mineral phase classification in forensic and geological applications without the need for extensive labeled real datasets.

Scanning electron microscopy (SEM) combined with energy dispersive spectroscopy (EDS) is widely used in geosciences for mineral phase classification. However, the lack of large, labeled datasets—especially for unpolished samples common in forensic pedology—limits the direct application of traditional supervised machine learning methods. In this study, we investigate the use of synthetic EDS data, generated using DTSA-II, to train advanced neural network architectures and evaluate their performance on real SEM–EDS measurements. We compare a U-Net baseline with a proposed 3D ResNet and a Transformer model. All models were trained on synthetic data and tested on real measurements of ten selected mineral phases. U-Net achieved accuracy of 63.7%, 3D ResNet reached 92.3%, and the Transformer model achieved the highest accuracy of 97.0%. These findings demonstrate that Transformer architectures can effectively generalize from synthetic to real EDS data, offering a promising way for accurate mineral phase classification in forensic and geological applications without the need for extensive labeled real datasets.

mineral phases; energy dispersive spectroscopy; neural networks; deep learning; transformers

mineral phases; energy dispersive spectroscopy; neural networks; deep learning; transformers

JONÁK, M.; DORAZIL, J.; MYŠKA, V.; BURGET, R.; KOTRLÝ, M.

2026

11.04.2025

979-8-3315-7675-2

Proceedings of the 17th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT)

International Congress on Ultra Modern Telecommunications and Workshops

Spojené státy americké

170

175

6