InSAR干涉图滤波方法比较

干涉条纹图滤波是合成孔径雷达干涉测量处理中不可缺少的重要环节。对干涉条纹图滤波的一个重要要求是在有效抑制噪声的同时尽可能地保持条纹的纹理信息。首先时干涉合成孔径雷达（InSAR）滤波方法进行分类，然后对基于局部统计自适应滤波中的4种算法：Goldstein滤波、INSAR干涉条纹图的复数空间自适应滤波、鲁棒加权圆周期滤波扣正余弦滤波法滤波进行了研究，每种算法都给出了真实数据的处理结果，以比较各种处理算法的优缺点。并且用5种定量的评价指标比较和评价上述4种滤波方法的优劣。

Comparison Among Filtering Methords of InSAR Interferogram

Filtering the interferometric fringes is one of the key processing procedures involved in synthetic aperture radar interferometry.The principal requirement of the filtering is to make the best maintenance to the fringe's texture while at the same time filter out the noises.In this paper,first,the filtering methods of radar interferogram were classified.Then,four adaptive interferogram filters,i.e.,the Goldstein filter,Interferometric phase complex adaptive filter,Robust weighted periodic pivoting filter,Sine or cosine interferogram filter,were studied.In order to make a comparison between the different filters, real data were used for experiments and results were provided.In addition,five quantitative measures for the quality of SAR interferograms were provided and used to evaluate the four interferogram filters.