中文摘要：

提出了一种自适应的InSAR干涉相位图滤波方法.干涉相位滤波算法的关键是独立同分布滤波样本的选取,该方法以相位变化最小为准则,根据区域增长策略,在当前滤波像素的邻域空间内自适应地选择滤波窗口,再在此窗口内进行复数平均得到滤波后的干涉相位值.以相位变化最小为准则进行滤波样本的选取,使其满足独立同分布的条件,从而提高干涉相位滤波的性能.仿真数据和实测数据的处理结果表明,该算法具有较好的滤波性能,能够充分保持干涉相位图像细节,并有效地去除噪声.

英文摘要：

This paper proposes a new adaptive noise suppressing method for interferometric phase images.The key point of noise filtering for interferometric phase images is the selection of the independent and identically distributed（i.i.d.） samples.The proposed method determines filtering samples with the criterion of the minimum phase differences according to the region-growing strategy,and then computes the filtered phase in the complex domain.Due to the minimum phase difference,the proposed method selects the filtering samples that obey the i.i.d.assumption more accurately,thus improving the filtering performance.Finally,experiments carried out on simulated and real data confirm that the proposed method possesses a better filtering performance and has the ability to suppres the noise effectively while maintaining the interferogram details adequately.