摘要
为充分挖掘矿区SAR影像丰富独特的空间信息,以SAR和SPOT5影像为数据源,基于目标成像知识理论,分析矿区典型目标成像特性,经几何校正等预处理,采用阈值分割、形态学滤波和基于特征融合等信息提取方法,建立基于特征知识信息提取的互补模式,对比分析SAR和SPOT5影像及融合后影像的解译能力.研究结果表明:该方法可使矿区露天采场、大型车辆、矿山道路网、复垦区、排土场、管线、建筑物等典型地物得到较好的解译.该成果在矿山监测中具有一定的应用价值和指导意义.
In order to fully tap the rich and unique spatial information of SAR images in mining areas, SAR and SPOT5 image was took to analysis the target imaging characteristics in mining area based on the theory, and image preprocessing such as geometric correction was conducted, and the method of threshold segmentation, morphological filtering and the feature information fusion method was used to extract information, and a complementary mode was established based on the characteristic knowledge information, and the interpretation ability of SAR,SPOT5 image and fused image were analyzed Lastly. The research result shows that feature information such as open stope mining areas, large vehicles, mining road nets, reclamation-areas, dumps, pipelines, buildings, etc. have got a better interpretation. The achievements can be applied to the mine monitoring.
出处
《辽宁工程技术大学学报（自然科学版）》
CAS
北大核心
2013年第6期768-772,共5页
Journal of Liaoning Technical University (Natural Science)
基金
辽宁省创新团队计划资助项目(2010-2012)
中国地质调查局矿山遥感调查与监测资助项目(121201122083)
