基于资源一号02D高光谱影像的典型蚀变矿物光谱参数模型及应用OA

Hyperspectral remote sensing technology,leveraging its unique advantage of integration of imagery and spectroscopy,has emerged as a pivotal technical approach in the domain of mineral resource exploration.To address the demand for rapid identification of typical alteration minerals in mineral resource surveys,this study employs the hyperspectral imagery acquired by the ZY-1 02D satellite over the Haidewula area in Qinghai as the data source.By integrating the spectral characteristics of uranium mineralization alteration minerals and the inherent noise properties of the imagery,spectral parameter models for typical alteration minerals—including sericite,chlorite,calcite,dolomite,and hematite—were constructed.Through a technical workflow encompassing image preprocessing,parameter model development,information extraction,and cartographic visualization,precise identification,spatial distribution of the target minerals in the study area were realized.This method effectively circumvents the threshold selection challenges associated with traditional spectral matching techniques and the sample dependency issue inherent in machine learning approaches,while simultaneously mitigating the interference of preprocessing noise on extraction outcomes.Experimental results demonstrated that the spatial distribution of the extracted minerals exhibited a high degree of consistency with field validation data from the study area.Therefore,this work provides an efficient and reliable remote sensing technical support for regional mineral resource surveys.