基于多源遥感数据的煤矿塌陷区快速动态监测方法的研究

Study on Fast Dynamic Monitoring Methods in Coal Mine Collapse Region Based on Multi-Source Remote Sensing Data

摘要: 本文主要利用Landsat TM/ETM+多时相遥感数据、SPOT纹理特性和高程DEM特征，采用多时相融合方法和主成分分析方法，得到变化的区域及属性特征；再利用GIS软件将近二十年来塌陷区变化情况分两期进行信息提取，从而获得1987年~2007年间矿区塌陷区变化情况，结合2013年DEM数据获得1987~2013年的塌陷区时空分布规律。结果表明，1987年到2000年塌陷区变化说明塌陷区面积逐年增加，开采重点主要在西部较新矿区；而2000年到2013年塌陷区面积变化较小，原因是期间塌陷区复垦为鱼塘，新村等。

Abstract: This paper aims to get mutative regions and attributes through the methods of multi-temporal fusion and principal component analysis which is based on Landsat TM/ETM multi-temporal remote sensing data, SPOT texture and DEM features. Then the information of nearly twenty years’ changes in collapse region is extracted in two periods by GIS software. Thus, combined with the 2013 DEM data, the temporal and spatial distribution rule of collapse region during 1987-2013 can be obtained. The results show that, collapse region changes from 1987 to 2000 illustrate that mining is mainly in Western newer mining area; but area change in collapse region from 2000 to 2013 is smaller because the collapse regions are reclaimed into fish ponds, villages and so on.