国产GF-5高光谱遥感在铀矿找矿中的应用——以甘肃省龙首山地区为例  

Application of GF-5 Satellite Hyperspectral Data in Uranium Exploration:A Case Study of Longshoushan Area in Gansu Province

作　　者：冯博[1,2] 段培新 程旭 卢辉雄 张恩[1,2] 汪冰 FENG Bo;DUAN Peixin;CHENG Xu;LU Huixiong;ZHANG En;WANG Bing(Airborne Survey and Remote Sensing Center of Nuclear Industry,Shijiazhuang,Hebei 050002,China;Hebei Key Laboratory of Airborne Survey and Remote Sensing Technology,Shijiazhuang,Hebei 050002,China;College of Earth Sciences,Hebei GEO University,Shijiazhuang,Hebei 050031,China)

机构地区：[1]核工业航测遥感中心,河北石家庄050002 [2]河北省航空探测与遥感技术重点实验室,河北石家庄050002 [3]河北地质大学地球科学学院,河北石家庄050031

出　　处：《铀矿地质》2022年第2期309-316,共8页Uranium Geology

基　　金：核工业地质局核工业航测遥感中心“航天高光谱遥感铀及多金属矿找矿技术研究”(编号:202004)资助。

摘　　要：GF-5观测卫星拥有着330个谱段的高光谱分辨率遥感数据,是我国2019年自主研发的首颗高光谱综合观测卫星。基于GF-5卫星开展的地质找矿应用的研究较少,文章选取了基岩裸露程度高、铀成矿地质条件优越的甘肃省龙首山铀成矿带作为GF-5高光谱遥感适宜性和可行性验证研究区,开展了数据预处理研究,构建了GF-5高光谱数据处理方法和流程,运用MNF算法、PPI算法,结合SAM光谱角填图技术,在研究区内共提取到包括钠长石、赤铁矿、高岭石、石英(硅化)、绿泥石和方解石在内的6种蚀变矿物。提取的蚀变矿物空间展布与区内铀矿(化)点、断裂构造和龙首山岩群分布特征关系密切,并通过开展地面光谱测量验证了蚀变矿物提取结果的可靠性。基于研究成果,丰富了GF-5高光谱的铀矿找矿工作思路,为国产GF-5高光谱遥感应用于地质找矿提供了参考。The GF-5 sensor has 330 spectral bands which is the first hyperspectral integrated observation satellite independently developed by China in 2019.However,there are few researches on the application of GF-5 satellite in geological prospecting.In this paper,the Longshoushan uranium metallogenic belt of Gansu province with high degree of outcrops and excellent uranium metallogenic geological conditions was selected as the GF-5 hyperspectral remote sensing suitability and feasibility verification research area.A Data preprocessing study was carried out in the research area,and The GF-5 hyperspectral data processing method and flow chart were constructed.Using MNF Algorithm,PPI Algorithm and Same Spectral Angle mapping technique,six altered minerals including Albite,hematite,kaolinite,quartz(silicified),chlorite and calcite were extracted in the study area.The spatial distribution of the extracted altered minerals is closely related to the distribution characteristics of uranium mineralization,fault structure and Longshoushan Group.The reliability of the extraction results of altered minerals was verified by ground spectral measurements.Based on the research results,this paper enriched the GF-5 hyperspectral method for uranium ore prospecting,and provided a reference for the application of GF-5 hyperspectral remote sensing in geological prospecting.

关 键 词：高分五号 航天高光谱 矿物识别 碱交代型铀矿 龙首山成矿带 

分 类 号：P627[天文地球—地质矿产勘探] P619.14[天文地球—地质学]