Machine Learning-Based Lithological Mapping from ASTER Remote-Sensing Imagery

Aug 3, 2026

Upload by Oliver

Accurate lithological mapping underpins geological surveys and mineral-resource exploration, while satellite remote-sensing offers cost-effective, widely applicable information on mineralized alteration zones. This study maps lithologies and minerals indirectly using machine-learning models trained on ASTER advanced spaceborne thermal emission and reflection radiometer imagery. Random forest, support vector machine, gradient boosting, extreme gradient boosting, and a deep-learning artificial neural network are assessed on the Sar-Cheshmeh copper mining area in southern Iran. Models are run in scenarios with extracted spectral features and with feature extraction omitted; feature importance filtering is also applied in the extraction workflow.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages27

Reading Time1 h 8 min

AccessFree to download

Introduction

Methodology

Machine-learning models

Feature extraction and feature importance filtering

Results and Accuracy Assessment

Lithological classes performance

Discussion and Conclusions

minerals Article Machine Learning-Based Lithological Mapping from ASTER Remote-Sensing Imagery Hazhir Bahrami 1, Pouya Esmaeili 2, Saeid Homayouni 1, Amin Beiranvand Pour 3, *, Karem Chokmani 1 and Abbas Bahroudi 2 Citation: Bahrami, H.; Esmaeili, P.; Homayouni, S.; Pour, A.B.; Chokmani, K.; Bahroudi, A. Machine Learning-Based Lithological Mapping from ASTER Remote-Sensing Imagery. Minerals 2024, 14, 202. [https://doi.org/](https://doi.org/)[ ](https://doi.org/)[10.3390/min14020202](10.3390/min14020202) Academic Editor: Huan Li Received: 21 December 2023 Revised: 11 February 2024 Accepted: 12 February 2024 Published: 16 February 2024 Copyright: © 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://](https://)[ ](https://)[creativecommons.org/licenses/by/](creativecommons.org/licenses/by/)[ ](creativecommons.org/licenses/by/)[4.0/](4.0/)) . 1 Centre Eau Terre Environnement, Institut National de la Recherche Scientiﬁque, Qu²bec, QC G1K 9A9, Canada; [hazhir.bahrami@inrs.ca](hazhir.bahrami@inrs.ca) (H.B.); [saeid.homayouni@inrs.ca](saeid.homayouni@inrs.ca) (S.H.); [karem.chokmani@inrs.ca](karem.chokmani@inrs.ca) (K.C.) 2 School of Mining Engineering, College of Engineering, University of Tehran, Tehran 1417935840, Iran; [poya.esmili@ut.ac.ir](poya.esmili@ut.ac.ir) (P.E.); [abbas.bahroudi@ut.ac.ir](abbas.bahroudi@ut.ac.ir) (A.B.) 3 Institute of Oceanography and Environment (INOS), Higher Institution Center of Excellence (HICoE) in Marine Science, Universiti Malaysia Terengganu (UMT), Kuala Nerus 21030, Terengganu, Malaysia * [Correspondence: beiranvand.pour@umt.edu.my](Correspondence: beiranvand.pour@umt.edu.my); Tel.: +60-9-6683824; Fax: +60-9-6692166 Abstract: Accurately mapping lithological features is essential for geological surveys and the exploration of mineral resources. Remote-sensing images have been widely used to extract information about mineralized alteration zones due to their cost-effectiveness and potential for being widely applied. Automated methods, such as machine-learning algorithms, for lithological mapping using satellite imagery have also received attention. This study aims to map lithologies and minerals indirectly through machine-learning algorithms using advanced spaceborne thermal emission andreﬂection radiometer (ASTER) remote-sensing data. The capabilities of several machine-learning (ML) algorithms were evaluated for lithological mapping, including random forest (RF), support vector machine (SVM), gradient boosting (GB), extreme gradient boosting (XGB), and a deep-learning artiﬁcial neural network (ANN). These methods were applied toASTER imagery of the Sar-Cheshmeh copper mining region of Kerman Province, in southern Iran. First, several spectral features that were extracted from ASTER bands were used as input data. Second, correlation coefﬁcients between the original spectral bands and features were extracted. The importance of the random forest features (RF's feature importance) was subsequently computed, and features with less importance were removed. Finally, the remained features were given to the models as input data in the second scenario. Accuracy assessments were performed for lithological classes in the study region, including Sar-Cheshmeh porphyry, quartz eye, late ﬁne porphyry, hornblende dike, granodiorite, feldspar dike, biotite dike, andesite, and alluvium. The overall accuracy results of lithological mapping showed that ML-based algorithms without feature extraction have the highest accuracy. The overall accuracy percentages for ML-based algorithms without conducting feature extraction were 84%, 85%, 80%, 82%, and 80% for RF, SVM, GB, XGB, and ANN, respectively. The results of this study would be of great interest to geologists for lithological mapping and mineral exploration, particularly for selecting app

FAQs

Why is lithological mapping important in geological surveys?

Lithological mapping supports bedrock surveys and mineral exploration by revealing the distribution and geological history of rock units and crust characteristics.

Which ASTER remote-sensing data and machine-learning methods are evaluated?

The study uses ASTER thermal emission and reflection imagery and evaluates random forest, support vector machine, gradient boosting, extreme gradient boosting, and a deep-learning artificial neural network.

What is the main finding about model accuracy with and without feature extraction?

ML-based algorithms without feature extraction achieve the highest overall accuracy, outperforming approaches that rely on extracted spectral features and feature-importance filtering.

Aug 3, 2026

Upload by Oliver