在分析复图像干涉相位对快速互相关算法配准精度影响的基础上,提出了一种快速区域互相关InSAR图像配准方法。该方法对图像的幅度谱进行区域相关操作,获得干涉相位空间角频率的粗估计,并对主图像进行干涉相位补偿。该算法通过上述步骤消除干涉相位变化对算法的影响实现了高精度配准。在仿真实验和实测实验中,通过与传统的快速区域互相关算法及最大频谱法进行对比,验证了本文算法的稳健性与有效性。
The complex image coregistration problem of interferometric synthetic aperture radar is addressed. By analyzing the effect of interferometric phase on the registration precision, a fast region correlation interferometric synthetic aperture radar (InSAR) image coregistration method is proposed. Firstly, it performs the fast region correlation technique on the magnitude-spectrum of two complex images, and then the spatial frequency of interferometric phase can be obtained coarsely. The estimated phase is compensated for the master image. The fast region correlation eoregistration method based on the interferometric phase effect elimination can achieve a high accuracy. The robustness of the proposed method can be verified by the simulated experimental results comparing with the correlation matching algorithm. The efficiency of the proposed method is demonstrated by actual data experimental comparing with the spec- tral maximum matching algorithm.