Abstract:

The mineral characteristic spectrum is in response to the mineral geochemical information, and they are closely linked with each other. These two kinds of information are to achieve precise correspondence on spatial position, and to achieve the coupling relationship on the different levels of both data according to the correlation of the two in genesis, for instance, data layer and decision-making (sign) level, and to improve the accuracy of extracting anomaly information. By analyzing the  remote sensing and geochemical information characteristics and formation mechanisms we can konw that both in space and on the causes they have a certain relevance. Use the inverse distance to a power, changed geochemical data into raster data structure and did spatial scale of the conversion and registration, selected four elements(Au,As,Sb,Bi) which related to the anomaly of the study area to weight fusion with remote sensing data, used the method of principal component analysis to process data, and then extracted the alteration information of Sonid Zuoqi area, Inner Mongolia, compared to using remote sensing data alone this method removed false anomaly points and get a more specific abnormal area. Through outdoor work, the field survey suggested that lots of surface rock alteration existed, proved the validity of the method.

Key words: spatial scale conversion, weighted fusion, principal component analysis, extraction of alteration information