一种综合地学数据的矿产资源潜力制图方法

A mineral resource potential mapping approach for synthesizing geological, geophysical, geochemical and remotely sensed data

研究适应信息化时代特征的矿产资源潜力制图新技术、新方法对推动矿产资源评价理论与技术的发展具有重要的意义.笔者把GIS技术、图像分类算法和空间统计学理论进行有机集成,在空间统计学的空间结构分析技术和遥感图像纹理分类算法的基础上,提出了一种以综合地学数据(地质、地球物理、地球化学和遥感图像数据等)为基本数据源的矿产资源潜力自动制图方法.该方法的技术流程为:①数据准备,即对地球物理和地球化学勘探数据进行预处理,生成一个物化遥综合图像文件;②图像空间结构性分析和纹理图像生成,以综合地学图像为研究对象,用空间统计学的结构分析技术研究地学数据综合图像的空间结构性,生成纹理图像;③纹理图像多元分类,用实验变差函数纹理分类方法对研究区进行多元分类,生成分类专题图;④分类后处理,用叠置分析修正空间分类结果,生成区域矿产资源潜力分布图.

It is a very important thing to investigate mineral resource potential mapping technologies and methods which suit 'Information Era'. Therefore, the authors propose a mineral resource potential mapping approach to synthesize geological, geophysical, geochemical, and remotely sensed data. This method is based on geostatistical theories and remotely sensed image texture classification algorithms. It needs to organically integrate GIS technologies, supervised classification algorithms and geostatistical theories. Its technological flowchart is as follows: ①Data preparation, i.e., preprocess geophysical and geochemical data to generate a comprehensive image file containing geophysical, geochemical and remotely sensed data; ②Image spatial structure analysis and texture image generation, i.e., analyze the spatial structures of geophysical, geochemical, and remotely sensed images using the spatial structure analysis techniques of geostatistics, to generate corresponding texture images; ③Multivariate classification of texture images, i.e., classify the study area by experimental semivariogram texture classification approaches to generate a classified thematic map; ④Post-classification process, i.e., modify classified thematic map with overlay analysis to generate a mineral resource potential map.