基于OLI和GF-2数据的锰矿化信息提取分析与对比--以新疆西昆仑玛尔坎苏锰矿带为例  

Analysis and Comparison of Extracting Manganese Mineralization Information from OLI and GF-2 Data:An Example of the Malkansu Manganese Mine in West Kunlun,Xinjiang

作　　者：李祥强 倪斌 彭欣 LI Xiangqiang;NI Bin;PENG Xin(Institute of Mineral Resources Research,China Metallurgical Geology Bureau,Beijing 101300)

机构地区：[1]中国冶金地质总局矿产资源研究院,北京101300

出　　处：《地质与勘探》2020年第5期1040-1049,共10页Geology and Exploration

基　　金：国家重点研发计划(编号:2016YFC0600210);国家自然科学基金(编号:41272366、41401515);基础性公益性地质矿产调查项目(编号:121201001000150013);中国冶金地质总局科技创新项目(编号:CMGB202001)联合资助。

摘　　要：本文在充分分析锰矿样品的实测波谱典型特征和矿化蚀变信息提取方法的基础上,基于Landsat OLI数据和国产GF-2数据影像,利用主成分分析法(PCA)对新疆西昆仑玛尔坎苏锰矿带进行了锰矿化异常信息提取,并对这两类数据的锰矿化信息提取结果进行了对比分析,进而讨论两类数据在锰矿化信息提取精度上的差异。研究表明:基于OLI和GF-2两类多光谱数据提取的锰矿化信息都取得了良好的效果,将已有的锰矿区与锰矿化点全部提取了出来,同时也提取出了未开采的锰矿化异常信息,验证了所采用信息提取方法的有效性;相比空间分辨率较低的OLI数据,高空间分辨率的国产GF-2遥感数据能够更好地区分锰矿化异常与阴影,更能精确地识别锰矿化异常的分布。因此,高空间分辨率多光谱数据在区分光谱特征差异较小的地物时具有较大优势,可为研究区下一步遥感找矿预测提供有利线索。This work builds on a full analysis of typical characteristics of measured spectra of manganese ore samples and extraction of mineralization and alteration information.Based on Landsat OLI and the domestic GF-2 multispectral data,we use the principal component analysis(PCA)method to extract the anomaly information of manganese mineralization in the Malkansu manganese ore belt in West Kunlun,Xinjiang,and compare the extraction results from the two types of data.Then we discuss their differences in the accuracy.Results show that the effects of information extraction from the both kinds of multispectral data,OLI and GF-2,are all fairly good.The existing manganese ore areas and manganese mineralization points are fully extracted,and the unmined manganese mineralization anomaly information is also extracted,which verifies the effectiveness of the information extraction method used.Compared with OLI data with low spatial resolution,the domestic GF-2 remote sensing data with high spatial resolution can better distinguish manganese mineralization anomalies and shadows,and identify the distribution of manganese mineralization anomalies more accurately.Therefore,multispectral data with high spatial resolution has a great advantage in distinguishing features of ground objectives with small spectral differences,which can provide more favorable clues for the next exploration in the study area.

关 键 词：OLI GF-2 锰矿化 主成分分析 玛尔坎苏锰带 西昆仑 新疆 

分 类 号：P627[天文地球—地质矿产勘探]