基于主成分分析和分形模型的ASTER蚀变异常信息提取  

The Extraction of Alteration Anomalies from ASTER Data Based on Principal Component Analysis and Fractal Model

作　　者：刁海[1] 张达[1] 狄永军[1] 王振[1] 王浩然[1] 熊光强[1] 

机构地区：[1]中国地质大学(北京)地球科学与资源学院,北京100083

出　　处：《国土资源遥感》2011年第2期75-80,共6页Remote Sensing for Land & Resources

基　　金：中国地质调查局地质调查项目"中蒙边境成矿带地质-地球物理综合剖面研究"(编号:1212010070505)

摘　　要：遥感蚀变异常信息在一定程度上可以反映出围岩蚀变情况。为了利用遥感数据提取围岩蚀变异常信息,通过分析蚀变矿物的波谱曲线,首先得出其在ASTER各波段的吸收、反射特征;然后选择特征明显的波段做主成分分析,并依据其光谱特征与特征向量的对应关系,确定出主要包含围岩蚀变信息的主分量;最后对该分量用分形模型计算其异常灰度的阈值,来提取蚀变异常信息。与已知矿点的围岩蚀变相比较,所提取的围岩蚀变异常信息与实际的蚀变情况比较吻合。To some extent, remote sensing alteration anomalies can reflect the alteration of wall rocks. The characteristics of absorption and reflection in ASTER can be obtained by analyzing the spectra of altered minerals. Some bands reflecting obvious characteristics can be calculated on the basis of principal component analysis, with the choice of the principal component of alteration information in accordance with the relationship between the spectra and the eigenvector. The thresholds of the anomalous gray are computed on the fractal model, and the extracted anomalies are well coincident with the known ore deposits.

关 键 词：主成分分析 分形模型 ASTER 蚀变异常提取 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置] P614[自动化与计算机技术—控制科学与工程]