Machine Learning Approaches for Real-Time Mineral Classification and Educational Applications

The main objective of the present study was to develop a real-time mineral classification system designed for multiple detection, which integrates classical computer vision techniques with advanced deep learning algorithms. The system employs three CNN architectures—VGG-16, Xception, and MobileNet V...

Full description

Saved in:
Bibliographic Details
Main Authors: Paraskevas Tsangaratos, Ioanna Ilia, Nikolaos Spanoudakis, Georgios Karageorgiou, Maria Perraki
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/4/1871
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849719791966748672
author Paraskevas Tsangaratos
Ioanna Ilia
Nikolaos Spanoudakis
Georgios Karageorgiou
Maria Perraki
author_facet Paraskevas Tsangaratos
Ioanna Ilia
Nikolaos Spanoudakis
Georgios Karageorgiou
Maria Perraki
author_sort Paraskevas Tsangaratos
collection DOAJ
description The main objective of the present study was to develop a real-time mineral classification system designed for multiple detection, which integrates classical computer vision techniques with advanced deep learning algorithms. The system employs three CNN architectures—VGG-16, Xception, and MobileNet V2—designed to identify multiple minerals within a single frame and output probabilities for various mineral types, including Pyrite, Aragonite, Quartz, Obsidian, Gypsum, Azurite, and Hematite. Among these, MobileNet V2 demonstrated exceptional performance, achieving the highest accuracy (98.98%) and the lowest loss (0.0202), while Xception and VGG-16 also performed competitively, excelling in feature extraction and detailed analyses, respectively. Gradient-weighted Class Activation Mapping visualizations illustrated the models’ ability to capture distinctive mineral features, enhancing interpretability. Furthermore, a stacking ensemble approach achieved an impressive accuracy of 99.71%, effectively leveraging the complementary strengths of individual models. Despite its robust performance, the ensemble method poses computational challenges, particularly for real-time applications on resource-constrained devices. The application of this methodology in Mineral Quest, an educational Python-based game, underscores its practical potential in geology education, mining, and geological surveys, offering an engaging and accurate tool for real-time mineral classification.
format Article
id doaj-art-e45c585c05104f8fa158afb99a8d5f32
institution DOAJ
issn 2076-3417
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-e45c585c05104f8fa158afb99a8d5f322025-08-20T03:12:04ZengMDPI AGApplied Sciences2076-34172025-02-01154187110.3390/app15041871Machine Learning Approaches for Real-Time Mineral Classification and Educational ApplicationsParaskevas Tsangaratos0Ioanna Ilia1Nikolaos Spanoudakis2Georgios Karageorgiou3Maria Perraki4Department of Geological Sciences, School of Mining and Metallurgical Engineering, National Technical University of Athens, 15773 Zografou, GreeceDepartment of Geological Sciences, School of Mining and Metallurgical Engineering, National Technical University of Athens, 15773 Zografou, GreeceSchool of Production Engineering and Management, Technical University of Crete (TUC), University Campus, Akrotiri, 73100 Chania, GreeceSchool of Electrical and Computer Engineering, Technical University of Crete (TUC), University Campus, Akrotiri, 73100 Chania, GreeceDepartment of Geological Sciences, School of Mining and Metallurgical Engineering, National Technical University of Athens, 15773 Zografou, GreeceThe main objective of the present study was to develop a real-time mineral classification system designed for multiple detection, which integrates classical computer vision techniques with advanced deep learning algorithms. The system employs three CNN architectures—VGG-16, Xception, and MobileNet V2—designed to identify multiple minerals within a single frame and output probabilities for various mineral types, including Pyrite, Aragonite, Quartz, Obsidian, Gypsum, Azurite, and Hematite. Among these, MobileNet V2 demonstrated exceptional performance, achieving the highest accuracy (98.98%) and the lowest loss (0.0202), while Xception and VGG-16 also performed competitively, excelling in feature extraction and detailed analyses, respectively. Gradient-weighted Class Activation Mapping visualizations illustrated the models’ ability to capture distinctive mineral features, enhancing interpretability. Furthermore, a stacking ensemble approach achieved an impressive accuracy of 99.71%, effectively leveraging the complementary strengths of individual models. Despite its robust performance, the ensemble method poses computational challenges, particularly for real-time applications on resource-constrained devices. The application of this methodology in Mineral Quest, an educational Python-based game, underscores its practical potential in geology education, mining, and geological surveys, offering an engaging and accurate tool for real-time mineral classification.https://www.mdpi.com/2076-3417/15/4/1871mineral classificationpre-trained CNN-based modelsGrad-CAM heatmapsquiz game
spellingShingle Paraskevas Tsangaratos
Ioanna Ilia
Nikolaos Spanoudakis
Georgios Karageorgiou
Maria Perraki
Machine Learning Approaches for Real-Time Mineral Classification and Educational Applications
Applied Sciences
mineral classification
pre-trained CNN-based models
Grad-CAM heatmaps
quiz game
title Machine Learning Approaches for Real-Time Mineral Classification and Educational Applications
title_full Machine Learning Approaches for Real-Time Mineral Classification and Educational Applications
title_fullStr Machine Learning Approaches for Real-Time Mineral Classification and Educational Applications
title_full_unstemmed Machine Learning Approaches for Real-Time Mineral Classification and Educational Applications
title_short Machine Learning Approaches for Real-Time Mineral Classification and Educational Applications
title_sort machine learning approaches for real time mineral classification and educational applications
topic mineral classification
pre-trained CNN-based models
Grad-CAM heatmaps
quiz game
url https://www.mdpi.com/2076-3417/15/4/1871
work_keys_str_mv AT paraskevastsangaratos machinelearningapproachesforrealtimemineralclassificationandeducationalapplications
AT ioannailia machinelearningapproachesforrealtimemineralclassificationandeducationalapplications
AT nikolaosspanoudakis machinelearningapproachesforrealtimemineralclassificationandeducationalapplications
AT georgioskarageorgiou machinelearningapproachesforrealtimemineralclassificationandeducationalapplications
AT mariaperraki machinelearningapproachesforrealtimemineralclassificationandeducationalapplications