Alteration Information Extraction In Vegetated Metallogenic Belts Of Southern Three Rivers Region,Southwest China

Alteration mineral information plays an important role in prospecting of mineral deposit. Compared with other geological information, alteration is usually weak information. And because of the interference of the vegetation and coverage, the extraction of the alteration information by remote sensing is very difficult by principal component analysis and band ratios which have been commonly applied to locate hydrothermal alteration zones related to metallic deposits in heavily vegetated areas. The paper mainly researched the mountainous region of Zhongdian, Yunnan province, China. A method for mapping of altered rocks in vegetated areas called ERVR (Enhanced Rock Vegetation Ratio) was developed and test at prong-hongshan. As reflection of vegetation is low in the shortwave infrared (SWIR), lithology and mineral is enhanced after Minimum Noise Fraction. In order to know means of physical significance of subset of MNF, mineral identifier was build. From the points of view about perspective of geology and physics, it discussed the mineralization conditions of porphyry copper mine and the spectral characteristics research of rock, altered wall rock, and altered ore. This is an important precondition and theoretic basis to extract remote sensing ore-controlling information. At the same time, by field-portable spectrum measurement instrument, the spectrums of typical deposits are collected, for example:quartz diorite porphyrite and beschtauite. And the mainly minerals are collected for USGS such as pyrite, muscovite, alunite, gadolinite. The technique turned target mineral spectral into pixels of remote sensing images as mineral identifier provides a potential approach for fast and effective to extract mineral alteration.The results support the role ERVR as very effective and robust image processing technique for mapping altered rocks and reducing interferences caused by vegetation in areas with green vegetation cover. Above study got the following conclusions and results:1) Standardization processing for field spectra as scale of image pixel. In order to get the regional spectrum of lithology rule of this area, filed and laboratory spectra was measured. And average and mean square error was calculated. The characteristic of landmark spectrum is not only the physical base of the remote sensing technical application but also the base of the quantificational analysis of remote sensing, and the study in landmark spectrums is the main content in the basic research of remote sensing. The characteristic of several typical landmark spectrums are set forth, the measure methods and its influential factors are introduced.2) Spectra of remote sensing image of vegetated areas was quantitatively analyzed. A method called Rock Vegetation Ration Index (RVI) was developed in order to evaluate identifiability of lithology for different ratio of rock and vegetation.3) A method of enhanced rock signal called as ERVR was developed. Firstly, the target mineral or lithology was accurate described. Secondly, mineral was enhanced in SWIR. Target mineral spectral was turned into pixels of images as mineral identifier provides a potential approach for fast and effective to extract mineral alteration from image data. The altered mineral was extracted from image covered vegetated areas.4) The technique turned target mineral spectral into pixels images was firstly suggested to our knowledge. It provides a potential approach for fast and effective to extract mineral alteration from ASTER multispectral data and Hyperion hyperspectral data. The role of mineral identifier is very effective not only for visual interpretation, but also the MNF. In this study, ERVR was applied to mapping altered mineral with ASTER and Hyperion in vegetated metallogenic belts of southern three riverside regions, southwest, China. The results support the role of ERVR as a very effective and robust image processing technique for extracting alteration minerals which are valuable for mineral explanation activities.By means of this research, preliminarily the method technique process on altered mineral information extraction in areas covered by vegetation is created. The result support the role of MNF and mineral identifier as very effective and robust image processing technique for mapping altered rocks tar and reducing interferences caused by vegetation in areas with green vegetation cover. Meanwhile, the technique will be ready technological ability for other areas covered vegetation applied widely in future.