中国非金属矿工业导刊Issue(2)：87-91,5.
基于GF-5高光谱数据的白云岩找矿应用研究
Application Research of Carbonate Rock Prospecting Based on GF-5 Hyperspectral Data--Taking the Tailya-Tianhu Area of the Eastern Tianshan Mountains as an Example
摘要
Abstract
Based on domestic GF-5 hyperspectral remote sensing data,a study on dolomite identification and prospecting prediction was carried out in the Weiya-Tianhu area of the eastern Tianshan Mountains in Xinjiang.Taking advantage of the well-exposed bedrock in the region,preprocessing steps such as radiometric calibration and atmospheric correction were applied.SWIR bands were selected for dimensionality reduction,and dolomite information was accurately extracted using a combination of optimal spectral selection and an integrated spectral shape-absorption position collaborative matching algorithm.Comparative analysis with mining areas such as Hengshan and southwestern Weiya showed that high-confidence identification zones highly coincide with the distribution of known ore bodies,while low-confidence zones correspond to marginal low-grade dolomite,verifying the reliability of the extraction results.Based on the analysis of metallogenic horizons and remote sensing identification results,a favorable prospecting area of 6.8 km2 was delineated.This study preliminarily confirms the effectiveness of GF-5 data in direct dolomite prospecting in shallow coverage areas and provides a reference for the application of hyperspectral remote sensing in non-metallic mineral exploration.关键词
高分五号/矿物识别/东天山尾亚—天湖地区/白云岩矿Key words
GF-5/mineral identification/east Tianshan Weiya-Tianhu area/dolomite deposits分类
天文与地球科学引用本文复制引用
刘春明,童锋,王俊峰,赵凯,岳连雄,王喜英,刘欢,孙科..基于GF-5高光谱数据的白云岩找矿应用研究[J].中国非金属矿工业导刊,2026,(2):87-91,5.基金项目
中煤航测遥感集团科技创新项目:基于航空高光谱"地形—光谱联合校正模型"的三维卷积神经网络岩性智能识别技术研究(编号:MHKJ-2025-7)项目资助. （编号:MHKJ-2025-7）
