湖北大冶铁矿矿山地质环境信息提取方法研究

Research on Geological Environmental Extraction Information of Daye Iron Mine in Hubei Province

湖北省黄石市素有"百里黄金地,江南聚宝盆"的美誉,但长期的矿产资源开发利用带来了严重的矿山地质环境问题。基于研究区矿山地质环境不同要素及其光谱特征,以"分层分类"思想为指导,利用TM数据构建了大冶铁矿矿山地质环境信息分层提取和决策树分类提取流程与方法体系。首先分析遥感数据统计特征,采用植被(水体)指数法将植被(水体)从地表环境基面中分离,其次运用主成分分析,结合光谱角法划分非植被、非水体和非居民地区域的"铁染异常"强度分布,最后在ENVI平台支持下采用决策树分类法对"铁染异常"区域作进一步划分。该方法能快速准确地将铁多金属矿的固体废弃、尾矿库、选矿场等分离,为矿山地质环境治理与恢复提供依据。

Huangshi City has been noted for its rich in mineral resources throughout the country, nicknamed ＂the treasure bowl in the southern part of Yangtze＂, but the mine resources development and utilization over a long period brought about serious problems in mining geological environment. Based on different factors and its spectral characteristic of mine geological environment, and under the guidance of hierarchical classification, TM data are used to set up the process and system of hierarchical extraction of mining geological environmental information and classified extraction of decision tree in Daye Iron Mine. In this paper, the characteristics of the remote sensing data statistics are firstly analyzed. The vegetation （body of water） index method is adopted to separate the vegetation （body of water） from the land surface. Secondly, using the main component analysis, spectrum classifications are combined to classify non--vegetation, non--body water and abnormal distribution of iron staining information in residential cision tree classification method is applied staining areas. The result shows that the waste, tailing pond, ore selection and so areas. Finally, on the platform of ENVI, the deto further divide the abnormal distribution in iron method can quickly and exactly separate the solid on from the iron polymetallic deposits.