摘要
以青海沟里地区为研究区,对区内资源一号02D(ZY1-02D)高光谱遥感影像进行预处理,充分分析了区内地质矿产资料,利用光谱角匹配技术提取典型蚀变矿物光谱样本数据,利用极端随机树算法开展矿化蚀变信息提取研究,并对提取效果进行评价。结果表明,提取的蚀变矿物信息分布位置与区内断裂构造展布特征以及已知矿床(点)的分布规律基本吻合,符合研究区成矿地质特征,说明结合光谱角匹配技术和极端随机树算法的蚀变矿物提取模型是一种高效的矿产勘查方法。
Taking Gouli area of Qinghai Province as the study area,we preprocessed the Ziyuan-1-02D(ZY1-02D)data hyperspectral remote sensing image in the area.Based on the analysis of geological and mineral data in the area,we extracted the spectral sample data of typical altered minerals based on spectral angle matching technology,used the extremely randomized tree algorithm to extract mineralization alteration information,and evaluated the extraction effect.The result shows that the distribution position of altered mineral information extracted by the model is basically consistent with the distribution characteristics of fault structures in the area and the distribution law of known deposits(points),which is in line with the metallogenic geological characteristics of study area,indicating this model is an efficient mineral exploration method.
作者
刘翰霖
李根军
林楠
杨雪松
LIU Hanlin;LI Genjun;LIN Nan;YANG Xuesong(Jilin Jianzhu University,Changchun 130118,China;Qinghai Geological Survey Institute,Xining 810008,China;Qinghai Kuntuo Remote Sensing Technology Service Co.,Ltd.,Xining 810007,China)
出处
《地理空间信息》
2023年第9期57-60,共4页
Geospatial Information
基金
国家自然科学基金资助项目(41702357)
青海省科技厅科技成果转化专项资助项目(2020-SF-150)。
