부산대학교 도서관

The core technology of Remote sensing mineralizing detection lies in extracting the information of ore-forming structure and altered mineral. The key of problem is comprehensive analysis and application of remote sensing image spectral information. Regarding Xinjiang Taxkorgan prefecture as the study area and choosing Multispectral ASTER remote sensing image data as data source, which can statistically analyze them including maximum, minimum, mean, standard deviation and covariance and so on. This method should explore and select the best band to extract the best band of Mineralized alteration anomaly information, applying principal component analysis to them based on basic geological data. Afterwards, it would assist mineral index (band ratio, operation) in the study area and mask processing model in extracting the mineralized alteration anomaly information of the prefecture. By contra posing the statistical analysis of the results of the prefecture and existing geological data, it could make out three metallogenic prospecting target areas and quantitatively predict the area of study on remote sensing mineralizing. It provides us a scientific solution that metallogenic prediction and evaluation should be carried out in the prefectures which is hard for traditional prospecting and have bad geological environment.