【摘要】 基于轨道数据的星载SAR影像直接地理定位技术是解决稀少控制条件下测图处理的重要手段之一,本文基于距离-多普勒(Range-Doppler,R-D)模型和地球物理模型,提出了一种由粗到精的TerraSAR影像直接地理定位方法。首先利用AIRGM算法对SAR影像目标进行粗定位,然后结合低分辨率DEM对SAR影像进行精定位,并利用陕西宝鸡地区高分辨率TerraSAR影像和SRTM数据对该算法的实用性进行了验证。结果表明,该算法可以在很大程度上提高其定位精度,能够满足该地区稀少控制条件下1∶50 000的地形图实际测图精度要求。更多还原

【Abstract】 SAR image direct geo-location based on the orbital data is one of important mapping means in some areas with sparse ground control points(GCPs).This paper proposed a kind of direct positioning method based on the Range-Doppler model and the geophysical model,which will execute two stages to realize the coarse-to-fine geo-location with TerraSAR images.First of all,using the AIRGM algorithm,it processed roughly to obtain the coarse positioning results of objects on SAR images,and then combining with lower-resolution DEM data,it acquired the precise positions.Moreover,it took a high-resolution TerraSAR-X image and SRTM data of Baoji area,in Shanxi province,as the test dataset to verify the practicality of this algorithm.The test results showed that the horizontal positioning accuracy supported with no GCPs was about 35m.More important,supported with only one ground control point,the horizontal positioning accuracy was about 8m.So it proved that this algorithm could improve the direct positioning accuracy to a great extent,and meet the practical requirements for 1:50 000 topographic mapping tasks in some areas with sparse GCPs.更多还原