面向城市道路目视识别的遥感图像滤波增强方法  

Remote Sensing Image Filtering Enhancement Method Based on the Visual Identification of City Roads

作　　者：吴乐芹[1,2] 姜春[1,3] 陈进栋[1] 王永梅[1] 邓南荣[1] 

机构地区：[1]广东省生态环境与土壤研究所,广东广州510650 [2]华南农业大学资源环境学院,广东广州510642 [3]中国科学院广州地球化学研究所,广东广州510640

出　　处：《地理空间信息》2015年第2期104-106,109,共4页Geospatial Information

基　　金：广东省科技计划项目(2010B060100068)

摘　　要：归纳与总结了遥感图像增强中的几种常用滤波方法。以广州城区作为试验区,Landsat TM影像作为数据源,分别运用纹理分析和卷积滤波对研究区域的遥感影像行增强处理,并对这些方法的影像增强效果进行质量评价。结果表明,高通滤波最适合于城市道路等线状地物的遥感图像增强。This paper briefly summarized several common filtering methods of remote sensing image enhancement.Taking the urban area of Guangzhou as study area and the Landsat TM image as data source,the methods of texture analysis and convolution filtering were used to enhance the remote sensing image of study area.By means of comparing the enhancement effect to assess the image quality,the results show that high pass filtering is the most suitable method of remote sensing image enhancement filtering for the visual identification of city roads and other linear features.

关 键 词：图像增强 滤波 纹理分析 卷积滤波 

分 类 号：P237.3[天文地球—摄影测量与遥感]