基于实时地形校正的InSAR图像匹配算法

针对侧视雷达/可见光图像匹配制导系统中由于雷达图像地形畸变引起的误匹配问题,提出了一种基于干涉合成孔径雷达（InSAR）的实时地形校正图像匹配算法。该算法以侧视雷达成像几何构象为基础,利用InSAR获取的实时地形数据对获取的SAR景象数据进行实时几何校正,生成无畸变的SAR景象数据,然后利用校正后的SAR景象数据与提前安装的可见光基准数据进行基于去均值归一化互相关模板的图像匹配。实验结果表明,通过实时地形校正,该景象匹配算法在复杂地形区域的匹配概率和匹配精度都大大优于传统SAR景象匹配算法,有效地提高了SAR图像匹配制导技术的适用性。

InSAR Image Matching Algorithm Based on Real-time Terrain Calibration

For the mismatch of side-looking radar/optical image due to geometric distortion of radar image,this paper presents an image matching algorithm based on real-time calibration by InSAR terrain.This algorithm is on the basis of geometric imaging model of side-looking radar.In this method,it firstly calibrate the real-time SAR scene by the real-time InSAR terrain data,and then match the calibrated real-time SAR scene and reference optical image set up in advance according to Zero-mean Normalized Cross-Correlation template.The experimental results show that the matching probability and matching accuracy in the complex terrain area of this algorithm is far superior to the traditional algorithm without InSAR terrain calibration.It has significantly improved the applicability of the SAR image matching guidance technology.