基于特征导向主成分分析遥感蚀变异常提取方法
Extraction of Alteration Anomaly Information by Feature-Based Principal Component Analysis from ASTER Data

为利用多光谱遥感数据提取蚀变异常信息,在分析蚀变矿物的先进星载热发射和反射辐射仪(advancedspaceborne thermalemissionandreflectionradiometer,ASTER)和影像短波近红外(visibleandnearIR-shortwave-lengthIR,VNIRSWIR) 谱带的特征光谱曲线的基础上,对传统的主成分分析法进行了改进,利用特征导向主成分分析法对辽宁兴城地区进行 矿物蚀变信息提取,成功的对该地区内的褐铁矿(Fe3+ )、绿泥石(Mg-OH 基团矿物)和高岭石(Al-OH 基团矿物)进行了蚀变 异常信息提取.通过实践验证和研究区地质资料表明,特征导向主成分分析法能够有效地提取蚀变信息并识别研究区内主要 矿物,可以为该区的成矿预测工作提供一定的依据.
In ordcr to cxtract thc altcration anomaly in」1ormation with multi-spcctra1 rcmotc scnsing data, this papcr improvcs the traditiona1 principa1 component analysis is improved and the minera1 alteration information of Xingcheng region of Liaoning using the method of the feature oriented principa1 component analysis is extracted, successfu11y for the alteration anomaly of the 1imonite(Fe3+), ch1orite(Mg-0H group minerals) and kaolinite(A1-〇H group minerals) based on the analysis of the charac- teristic curve of canopy spectra1 in VNIR-SWIR(visible and near IR-short wave-1ength IR) band of ASTER(advanced space- borne therma1 emission and reflection radiometer) data about alteration minera1.Field practice and geo1ogica1 data of research ar- ea indicate that the method above can effectively extract the alteration information and identify the mainly minera1 of research area, which can provide a certain basis for the meta11ogenic prediction