高分辨率遥感数据处理及其铀矿地质应用

高光谱和高空间分辨率遥感技术在铀矿地质勘查中具有重要作用,遥感数据挖掘与知识发现、新方法研究则是各种铀成矿要素光谱信息和空间特征识别的关键技术手段。本文基于曲波变换技术,开发了卫星高光谱数据（Hyperion）和高空间分辨率（SPOT5）的影像融合方法,融合影像在数据噪声消除、信息增强与信息提取方面应用效果显著;利用离散小波变换技术,高精度提取了主要铀成矿要素的光谱参数,建立了典型四价和六价铀矿物的光谱识别谱系。另外,通过钻孔岩心和探槽工程的高光谱编录、地质编录和物探编录,建立了典型蚀变带的物谱关系模型,重点分析了赤铁矿化蚀变（Fe3＋）光谱特征与铀含量之间的相关关系。上述遥感数据处理新技术新方法在桃山地区铀矿地质勘查中取得了理想效果。

High resolution remote sensing image processing technology and its application to uranium geology

Hyperspectral and high spatial resolution remote sensing technology take important role in uranium geological application,data mining and knowledge discovery methods are key to character spectral and spatial information of uranium mineralization factors.Based on curvelet transform algorithm,this paper developed the image fusion technology of hyperspectral（Hyperion） and high spatial data（SPOT5）,and results demonstrated that fusion image had advantage in denoising,enhancing and information identification.Used discrete wavelet transform,the spectral parameters of uranium mineralization factors were acquired,the spectral identification pedigrees of typical quadrivalence and hexavalence uranium minerals were established.Furthermore,utilizing hyperspectral remote sensing observation technology,this paper developed hyperspectral logging of drill cores and trench,it can quickly process lots of geological and spectral information,and the relationship between radioactive intensity and abnormal spectral characteristics of Fe3＋ was established.All those provided remote sensing technical bases to uranium geology,and the better results have been achieved in Taoshan uranium deposits in south China.