不同时空条件下的高分三号SAR图像立体定位方法  

A stereo positioning method of GaoFen-3 SAR images under different time and space conditions

作　　者：李丹 刘佳音 LI Dan;LIU Jiayin(Key Laboratory of Technology in Geo-Spatial Information Processing and Application Systems of CAS,Aerospace Information Research Institute,Chinese Academy of Sciences,Beijing 100094,China;School of Electronic,Electrical and Communication Engineering,University of Chinese Academy of Sciences,Beijing 100049,China)

机构地区：[1]中国科学院空天信息创新研究院中国科学院空间信息处理与应用系统技术重点实验室,北京100094 [2]中国科学院大学电子电气与通信工程学院,北京100049

出　　处：《中国科学院大学学报（中英文）》2021年第4期519-523,共5页Journal of University of Chinese Academy of Sciences

基　　金：国家重点研发计划(2017YFB0502901)资助。

摘　　要：介绍一种较为可靠的高分三号合成孔径雷达(synthetic aperture radar,SAR)图像立体定位方法,即便立体像对时空条件差异较大,该方法也能达到较高的定位精度。首先根据斜距多普勒模型建立立体定位方程组,利用多项式拟合法解算卫星状态矢量,最后利用牛顿法求解方程组。利用高分三号卫星在北京、苏州、上海3个地区、成像时间各不相同且分辨率在5 m以内的7景SAR图像进行立体定位实验,以Google Earth上的实际位置为基准,实验得到的所有目标点的平面误差均在30 m内,高程误差均在34 m内,证明即使立体像对成像时间间隔较长,也可以得到精度较高的目标点三维信息,且滑动聚束模式下得到的立体像对定位精度最高。This paper introduces a reliable stereo positioning method of GaoFen-3 synthetic aperture radar(SAR)images.The method can achieve high positioning accuracy even if the spatial and temporal conditions of the stereo image differ greatly.Firstly,the stereo positioning equations are established according to the range-Doppler model,and then the polynomial fitting method is used to obtain the state vectors of the satellite.Finally,the equations are solved by Newton’s method.In this paper,stereo location experiments are carried out with 7 SAR images of Beijing,Shanghai,and Suzhou with different imaging times.The resolutions of these images are within 5 m.Verification based on Google Earth image shows that the plane errors of all the targets are within 30 m,and the height errors are within 34 m.It is proved that even if the imaging time intervals of the stereo SAR image pairs are long,the 3D information with high accuracy can also be obtained.Besides,the localization accuracy of the SAR image pair from spotlight mode is the highest.

关 键 词：SAR立体图像对 高分三号 斜距多普勒模型 多项式拟合法 牛顿法 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]