Sitemap Contact Us
닫기
Search
Advanced Search +
닫기
Quick links
Archives Most Cited Most Read Review My Read Go to E-submission

Research Paper

Split Viewer

Econ. Environ. Geol. 2022; 55(2): 197-207

Published online April 30, 2022

https://doi.org/10.9719/EEG.2022.55.2.197

© THE KOREAN SOCIETY OF ECONOMIC AND ENVIRONMENTAL GEOLOGY

Detection of Titanium bearing Myeonsan Formation in the Joseon Supergroup based on Spectral Analysis and Machine Learning Techniques

Chanhyeok Park1, Jaehyung Yu2,*, Min-Kyu Oh1, Gilljae Lee3, Giyeon Lee2

Author Info. +

1Department of Astronomy, Space Science, & Geology, Chungnam National University
2Department of Geological Sciences, Chungnam National University
3Rare Metal Ore Research Center, Korea Institute of Geoscience and Mineral Resources

Correspondence to : *Corresponding author : jaeyu@cnu.ac.kr

Received: April 8, 2022; Revised: April 26, 2022; Accepted: April 26, 2022

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided original work is properly cited.

Abstract

This study investigated spectroscopic exploration of Myeonsan formation, the titanium(Ti) ore hostrock, in Joseon supergroup based on machine learning technique. The mineral composition, Ti concentration, spectral characteristics of Myeonsan and non-Myeonsan formation of Joseon supergroup were analyzed. The Myeonsan formation contains relatively larger quantity of opaque minerals along with quartz and clay minerals. The PXRF analysis revealed that the Ti concentration of Myeosan formation is at least 10 times larger than the other formations with bi-modal distribution. The bi-modal concentration is caused by high Ti concentrated sandy layer and relatively lower Ti concentrated muddy layer. The spectral characteristics of Myeonsan formation is manifested by Fe oxides at near infrared and clay minerals at shortwave infrared bands. The Ti exploration is expected to be more effective on detection of hostrock rather than Ti ore because ilmenite does not have characteristic spectral features. The random-forest machine learning classification detected the Myeonsan fomation at 85% accuracy with overall accuracy of 97%, where spectral features of iron oxides and clay minerals played an important role. It indicates that spectral analysis can detect the Ti host rock effectively, and can contribute for UAV based remote sensing for Ti exploration.

Keywords titanium, Myeonsan formation, spectral analysis, machine learning, ore detection

분광분석과 기계학습기법을 활용한 조선누층군 타이타늄 함유 면산층 탐지

박찬혁1 · 유재형2,* · 오민규1 · 이길재3 · 이기연2

1충남대학교 우주·지질학과
2충남대학교 지질환경과학과
3한국지질자원연구원 희소금속광상연구센터

요 약

본 연구는 조선누층군 내 타이타늄 광체의 모암이 되는 면산층 암석을 기계학습기법을 분광분석 결과에 적용하여 탐지하였다. 이를 위해 면산층과 타 층들의 구성 광물을 파악하고, 타이타늄 함량을 측정하였으며, 전자기파 반응 특성을 분석하였다. 면산층은 다른 층들에 비해 불투명 광물을 많이 함유하고, 석영 입자와 점토광물로 구성된다. X선 형광분석 결과, 면산층의 평균 타이타늄 함량은 타 층들에 비해 최소 10배 이상의 타이타늄 함량을 보이며 낮은 함량군과 높은 함량군의 다봉분포를 갖는다. 이는 면산층 내의 타이타늄이 함유되는 사질과 이질이 교호 반복되는데 사질 부분은 이질 부분보다 타이타늄의 함량이 상대적으로 높기 때문이다. 분광분석 결과, 면산층은 산화철의 흡광 특성이 근적외선 영역에서, 점토광물에 의한 흡광 특성이 단파적외선 영역에서 관찰되며, 풍화면의 경우 점토광물 특성이 보다 강해지는 경향을 보인다. 타이타늄 광화대의 탐지는 티탄철석 자체의 분광 특성이 특징적이지 않아 광체를 탐지의 대상으로 보기보다는 모암인 면산층을 탐지하는 것이 적절할 것으로 생각된다. 랜덤포레스트 기계학습 기법을 이용한 면산층의 탐지 정확도는 84%, 전체정확도 97%를 보였으며, 산화철의 분광 특성과 점토광물 분광 특성이 가장 중요한 역할을 하는 것으로 분석되었다. 이는 분광 특성이 타이타늄 모암인 면산층 암석을 효율적으로 탐지할 수 있음을 지시하고, 확대 적용 될경우 무인항공기반 타이타늄 광체 탐사에 적용할 수 있을 것으로 기대한다.

주요어 타이타늄, 면산층, 분광분석, 기계학습, 광체탐지

Article

Research Paper

Econ. Environ. Geol. 2022; 55(2): 197-207

Published online April 30, 2022 https://doi.org/10.9719/EEG.2022.55.2.197

Copyright © THE KOREAN SOCIETY OF ECONOMIC AND ENVIRONMENTAL GEOLOGY.

Detection of Titanium bearing Myeonsan Formation in the Joseon Supergroup based on Spectral Analysis and Machine Learning Techniques

Chanhyeok Park1, Jaehyung Yu2,*, Min-Kyu Oh1, Gilljae Lee3, Giyeon Lee2

1Department of Astronomy, Space Science, & Geology, Chungnam National University
2Department of Geological Sciences, Chungnam National University
3Rare Metal Ore Research Center, Korea Institute of Geoscience and Mineral Resources

Correspondence to:*Corresponding author : jaeyu@cnu.ac.kr

Received: April 8, 2022; Revised: April 26, 2022; Accepted: April 26, 2022

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided original work is properly cited.

Abstract

This study investigated spectroscopic exploration of Myeonsan formation, the titanium(Ti) ore hostrock, in Joseon supergroup based on machine learning technique. The mineral composition, Ti concentration, spectral characteristics of Myeonsan and non-Myeonsan formation of Joseon supergroup were analyzed. The Myeonsan formation contains relatively larger quantity of opaque minerals along with quartz and clay minerals. The PXRF analysis revealed that the Ti concentration of Myeosan formation is at least 10 times larger than the other formations with bi-modal distribution. The bi-modal concentration is caused by high Ti concentrated sandy layer and relatively lower Ti concentrated muddy layer. The spectral characteristics of Myeonsan formation is manifested by Fe oxides at near infrared and clay minerals at shortwave infrared bands. The Ti exploration is expected to be more effective on detection of hostrock rather than Ti ore because ilmenite does not have characteristic spectral features. The random-forest machine learning classification detected the Myeonsan fomation at 85% accuracy with overall accuracy of 97%, where spectral features of iron oxides and clay minerals played an important role. It indicates that spectral analysis can detect the Ti host rock effectively, and can contribute for UAV based remote sensing for Ti exploration.

Keywords titanium, Myeonsan formation, spectral analysis, machine learning, ore detection

분광분석과 기계학습기법을 활용한 조선누층군 타이타늄 함유 면산층 탐지

박찬혁1 · 유재형2,* · 오민규1 · 이길재3 · 이기연2

1충남대학교 우주·지질학과
2충남대학교 지질환경과학과
3한국지질자원연구원 희소금속광상연구센터

Received: April 8, 2022; Revised: April 26, 2022; Accepted: April 26, 2022

요 약

본 연구는 조선누층군 내 타이타늄 광체의 모암이 되는 면산층 암석을 기계학습기법을 분광분석 결과에 적용하여 탐지하였다. 이를 위해 면산층과 타 층들의 구성 광물을 파악하고, 타이타늄 함량을 측정하였으며, 전자기파 반응 특성을 분석하였다. 면산층은 다른 층들에 비해 불투명 광물을 많이 함유하고, 석영 입자와 점토광물로 구성된다. X선 형광분석 결과, 면산층의 평균 타이타늄 함량은 타 층들에 비해 최소 10배 이상의 타이타늄 함량을 보이며 낮은 함량군과 높은 함량군의 다봉분포를 갖는다. 이는 면산층 내의 타이타늄이 함유되는 사질과 이질이 교호 반복되는데 사질 부분은 이질 부분보다 타이타늄의 함량이 상대적으로 높기 때문이다. 분광분석 결과, 면산층은 산화철의 흡광 특성이 근적외선 영역에서, 점토광물에 의한 흡광 특성이 단파적외선 영역에서 관찰되며, 풍화면의 경우 점토광물 특성이 보다 강해지는 경향을 보인다. 타이타늄 광화대의 탐지는 티탄철석 자체의 분광 특성이 특징적이지 않아 광체를 탐지의 대상으로 보기보다는 모암인 면산층을 탐지하는 것이 적절할 것으로 생각된다. 랜덤포레스트 기계학습 기법을 이용한 면산층의 탐지 정확도는 84%, 전체정확도 97%를 보였으며, 산화철의 분광 특성과 점토광물 분광 특성이 가장 중요한 역할을 하는 것으로 분석되었다. 이는 분광 특성이 타이타늄 모암인 면산층 암석을 효율적으로 탐지할 수 있음을 지시하고, 확대 적용 될경우 무인항공기반 타이타늄 광체 탐사에 적용할 수 있을 것으로 기대한다.

주요어 타이타늄, 면산층, 분광분석, 기계학습, 광체탐지

    Fig 1.

    Download
    Figure 1.Schematic sedimentary log of the Cambrian-Ordovician Taebaek Group. Modified after Lee et al. (2016).
    Economic and Environmental Geology 2022; 55: 197-207https://doi.org/10.9719/EEG.2022.55.2.197

    Fig 2.

    Download
    Figure 2.Geological map of the Seokgaejae area. Maps modified after GICTR (1962), Choi et al. (2004), and Chough (2013).
    Economic and Environmental Geology 2022; 55: 197-207https://doi.org/10.9719/EEG.2022.55.2.197

    Fig 3.

    Download
    Figure 3.Photographs of rock samples used for this study showing weathered and fresh surface; a) Myonsan Formation, b) Dumugol Formation, c) Hwajeol Formation, d) Myeonsan Formation, e) Makgol Formation, and f) Daegi Formation.
    Economic and Environmental Geology 2022; 55: 197-207https://doi.org/10.9719/EEG.2022.55.2.197

    Fig 4.

    Download
    Figure 4.Photomicrographs of the thin-sections for the rock samples. (a) Myeonsan Formation containing very fine sand-sized quartz (Qz), detrital heavy minerals (Hm), and clay minerals. (b) Dolomite crystals and a fossil fragment composed of calcite from the Daegi Formation (c) Limestone-shale couplet in the Hwajeol Formation. The light-gray lime mudstone is interbedding with brownish shale. (d) The Dongjeom Formation consisting of mainly monocrystalline quartz (Mono-qz) and minor polycrystalline quartz (Poly-qz). The monocrystalline quartz grains commonly show overgrowth. (e) The Dumugol Formation shows intercalation of lime mudstone and wacke- to packstone. Euhedral dolomite (Dl) crystals are abundant in the wacke- to packstone. (f) Massive micrite of the Makgol Formation.
    Economic and Environmental Geology 2022; 55: 197-207https://doi.org/10.9719/EEG.2022.55.2.197

    Fig 5.

    Download
    Figure 5.Reflectance spectra(a) and Hull quotient continuum removal spectra(b) of Joseon supergroup and USGS library spectra of caly and carbonate minerals.
    Economic and Environmental Geology 2022; 55: 197-207https://doi.org/10.9719/EEG.2022.55.2.197

    Fig 6.

    Download
    Figure 6.Variable importance for detection of Myeonsan Formation derived from Gini decay.
    Economic and Environmental Geology 2022; 55: 197-207https://doi.org/10.9719/EEG.2022.55.2.197

    Table 1 . Criteria for random forest Classification.

    TargetClassification Criteria
    Myeonsan FormationClass 1: Myeonsan Formation
    Class 2: Non-Myeonsan Formation

    Table 2 . Minimum, maximum, mean, and standard deviation of titanium content in Joseon Supergroup samples from Seogaejae.

    Formation
    DongjeomDumugolMakgolMyeonsanDaegiHwajeol
    Min (%)0.0690.0660.0530.5070.0560.077
    Max (%)0.2330.3280.13721.3610.0900.637
    Mean (%)0.1900.1580.0824.3460.0710.385
    Standard deviation0.0520.0660.0186.6730.0120.110

    Table 3 . Accuracy assessment of random forest classification for detection of Ti bearing Myeonsan formation result.

    Classification criteriaPrediction
    Case 1Non-Myeonsan formationMyeonsan FormationError(%)Sum of row
    Non-Myeonsan formation18900.0189
    Myeonsan Formation73417.041
    Sum of column196343.0230

    KSEEG
    Dec 31, 2024 Vol.57 No.6, pp. 665~835
    Current Issue Archives

    Stats or Metrics

    Share this article on

    • kakao talk
    • line

    Related articles in KSEEG

    Economic and Environmental Geology

    pISSN 1225-7281
    eISSN 2288-7962
    qr-code Download