SAR影像的几何精纠正

SAR影像几何纠正中的一个重要问题是建立其构像模型。在阐述F．Leberl模型的基本原理和方法的基础上，针对星载SAR成像处理过程中多普勒中心频率不为零的现象，采用多项式拟和多普勒中心频率的方法，修正零多普勒条件方程，建立起既适合星载又适合机载SAR传感器的F．Leberl构像模型。机载和星载SAR影像的实验结果表明，文中提出的改进算法能显著地提高F．Leberl模型对星载SAR影像的定位精度，同时也适合于对机载SAR数据的处理，并且解算过程稳定，收敛很快，精度较高。

Precise Rectification of SAR Images

With the development of SAR(Synthetic Aperture Radar) remote sensing techniques,SAR images have been widely applied in a lot of fields.But because of its side-looking character,we have to rectify the images into orthoimages before we use them.Building mathematical models is an important step in SAR rectification.In this paper we first recapitulate some mathematical models used for SAR image rectificaion,including Soviet Model,Polynomial Model and G.Konecny Model.The characteristics of these models are analyzed.Then we introduce the F.Leberl Model in detail,which is based on the distant condition and the zero-Doppler condition.In airborne condition the range direction is perpendicular to the flight track so the Doppler center frequency is zero.But in spaceborne condition,the Doppler center frequency does not always equal to zero,so the zero-Doppler condition needs to be corrected.In this article we use a polynomial to substitute the value of the Doppler center.By this way we corrected the zero-Doppler condition.In the experiments we used both the original F.Leberl Model and the corrected model to validate the precision of our model.From the results we come to the conclusion that the new model is acceptable to both the airborne SAR and the spaceborne SAR,at the same time it can get higher precision than the original model in spaceborne condition.