基于LTS稳健初值的选权迭代法在SAR图像配准中应用研究

分析SAR图像配准联系点偏移量粗差对配准精度的影响，采用基于LTS（least trimmed squares）稳健初值的选权迭代法剔除偏移量粗差。以南通地区ENVISAT ASAR数据为例，选取干涉图条纹质量和相干系数总体分布作为评价指标，对该粗差剔除方法进行实验。结果表明，该算法能有效去除偏移量粗差，消除偏移量粗差对配准精度的影响，提高SAR配准的精度。

SAR Registration Based on Selecting Weight Iteration Method with Initial Value by LTS

In this paper, we analyze the effect of offset outliers on the accuracy of SAR registration. In order to solve this problem, we adopt an algorithm that combines the LTS and selecting weight iteration. Then, we use ENVISAT ASAR data in Nantong as an example and choose the quality of the interferogram fringe and the distribution of the coherence as the index to test the robustness of our method. The results indicate that the proposed method is able to reject the offset outliers effectively and eliminates their effect on the accuracy of SAR registration. Thus, our method can improve the accuracy of SAR registration reliably.