摘要
滤波滑窗的选取是SAR图像相干斑滤波的关键问题之一,从SAR图像中均匀区域的幅度和空间特性出发,提出了一种基于置信区间与形态学重构的自适应滑窗方法。首先在初始滑窗内进行强度筛选,依据初筛结果的异质程度自适应调整置信区间,得到在幅度上连续的初始自适应窗;接着以初始自适应窗为掩膜,通过空间邻接约束进行形态学重构,进一步得到在空间上连续的、形状任意的自适应滑窗。实验结果表明,与固定窗法和改进的Sigma方法相比,该方法能够获得更准确的滤波滑窗。
Filter window selection is one of the key issues in SAR image despeckling.An adaptive windowing method for speckle reduction is proposed,which is based on the combination of confidence interval and morphological reconstruction.Pixel selecting is first proceeded in a fixed window based on the radiometric confidence interval.A confidence interval is chosen adaptively according to the homogeneity facts of current window.Then a region adjacency constrain is carried out by morphological reconstruction to refine the window to a radiometric and spatial continuous window with arbitrary shape.The experiment results show that the proposed adaptive windowing method performs better speckle reduction and structure preservation than the box filter and improved Sigma filter.
出处
《遥感技术与应用》
CSCD
北大核心
2011年第3期315-321,共7页
Remote Sensing Technology and Application
关键词
相干斑滤波
自适应滑窗
置信区间
形态学重构
Speckle reduction
Adaptive windowing
Confidence interval
Morphological reconstruction
