Application Of Fractal Theories In Alteration Information Extraction Of East Tianshan Mineral District Based On Remote Sensing Images

The east Tianshan area,being an important prospecting area in Xinjiang is rich in diverse mineral resources.In recent years,the prospecting research work for this area has been increasing.The use of remote sensing images,combined with a variety of methods to extract alteration information for mineral resources prediction work,has a certain representative.However,the traditional information extraction methods show certain limitations.The extracted alteration anomaly results generally contain a large amount of background information components.Therefore,removing irrelevant background information and obtaining effective anomaly information are the keys to using remote sensing prospecting.There are always irregularities in things in nature.Fractal theories can reveal self-similar behind the phenomenon by analyzing the seemingly irregular phenomenon.Therefore,the fractal methods can then be used to analyze the remote sensing alteration information.Starting from the metallogenic theory,this study selects the most typical types of mineralization in the study area—porphyry copper deposit,magmatic copper-nickel deposit and marine volcanic sedimentary iron deposit as the research objects,with multi-spectral remote sensing technology.The fractal theories are applied to the research of remote sensing alteration information extraction.The feasibility and a flow chart of fractal theory applied in alteration information identification are explored.It shows that the fractal method can obtain better results in the identification of remote sensing alteration information than traditional methods,such as ratio method and principal component analysis.The research contents and results of this paper are as follows:1.This paper applies terrestrial satellite image data,and analyzes the spectral curves of different wavelengths of remote sensing images according to the different absorption and reflection characteristics of altered minerals or rocks.A technical flow consisting masking,principal component analyzing,Spectrum-Area(S-A)filtering and alteration information grading is currently proposed.The process of extracting the information of remote sensing alteration anomaly formed by the steps provides clues for remote sensing prospecting in the baring area.2.This paper compares and analyzes the two commonly used fractal models—the energy spectrum-area model(S-A model)and the Concentration-Area(C-A model).The former is applied to the separation of remote sensing alterationanomaly and background information,and the latter is applied to the research of remote sensing alteration anomaly classification.Based on the interpretation of remote sensing images,the fault structure is extracted.According to the characteristics of three typical deposits,metallogenic regularity and ore-controlling structure of Huangshan,Tuwu and Yamansu,the relationship between the geological connotation represented by S-A energy spectrum and different deposit types is studied.3.In the classification of alteration information,this paper uses the threshold method and Concentration-Area(C-A)method to obtain the lower limit of abnormality,respectively.Based on the results,the classification maps of alteration information of different mining areas are drawn,and the existing mineral deposits are superimposed for field verification and comparative analysis.The results show that the threshold obtained by C-A method has stronger geological significance.