【摘要】 为了保持多极化SAR图像的极化特性,协方差矩阵中的每一个元素都必须在相同的局部同质区内进行独立的滤波处理。大多数滤波方法都采用了固定大小的滑动窗技术,导致其同质区域的大小不能自适应于局部纹理的变化,因此,很难解决保留结构特征和平滑斑点噪声这对矛盾。要解决该问题,可采用大小可变的滑动窗来搜索局部的最大同质区,最大同质区的大小能自适应局部纹理的变化。实验结果表明,该方法能有效地保持图像的空间结构特征和极化特性,同时也能获得较好的斑点噪声抑制效果。更多还原

【Abstract】 To preserve the polarimetric properties of polarimetric SAR image,each element of the covariance matrix should be filtered independently in the same homogenous area.Most of present filtering methods adopt a fixed-size sliding-window technology,which causes the size of homogenous area is not adaptive to the local texture,and hence it’s very hard to make a trade-off between the preservation of spatial structure features and the smoothness of speckle noise.Aiming at this problem,a flexible-size sliding-window is used to search the maximum-size homogenous area,and consequently the size of homogenous area can be adaptive to the local texture.The experimental results illustrated that the proposed method can effectively preserve the polarimetric properties as well as the spatial structure features,meanwhile it has a remarkable despeckling ability.更多还原