【作者】 王霞；

【导师】 吴艳；

【作者基本信息】 西安电子科技大学 ， 电路与系统， 2006， 硕士

【摘要】 抑制SAR图像斑点技术的研究一直是SAR成像处理与图像分析中的一个重要课题。本文在小波域模型的基础上,提出了两种SAR图像降斑算法,仿真结果表明两种算法取得了预期效果。提出了基于平稳小波变换贝叶斯模糊萎缩的SAR图像斑点抑制算法。分析了SAR图像在平稳小波变换域中的统计模型,推导出基于贝叶斯估计的信号最小均方误差(MMSE)的模糊萎缩因子。籍此再根据相邻尺度间小波系数的相关性,采用分区域模糊萎缩思想,很好地得到无斑点真实信号小波系数的估计。提出了小波域隐马尔可夫混合模型的SAR图像降斑算法。该模型考虑了小波系数的持续性和聚类性,分别用隐马尔可夫树(HMT)和隐马尔可夫链(HMC)刻画。并且采用的半树模型提高了HMT算法的运行速度,HMC可以更精确地估计该模型根节点的初始值。该算法融合贝叶斯MMSE抑制噪声技术,很好地得到无斑点真实信号小波系数的估计。更多还原

【Abstract】 The research of synthetic aperture radar (SAR) images despeckling methods has always been a difficult matter in the regions of SAR imaging and its postprocessing. Based on wavelet, two SAR image despeckling methods are proposed in this paper. And experimental results show that they have achieved good performance.One efficient despeckling method is proposed based on stationary wavelet transform (SWT) for SAR images. The statistical model of wavelet coefficients is analyzed. A fuzzy shrinkage factor is derived based on the minimum mean square error (MMSE) criteria with Bayesian estimation. In the case above, the ideas of regions divided and fuzzy shrinkage are adopted according to the intrascale dependencies of the wavelet coefficients. The noise-free wavelet coefficients are estimated finely.The other efficient despeckling method is proposed based on wavelet. The algorithm accounts for the clustering and the persistence of wavelet coefficients, which are characterized by hidden markov chain (HMC) and hidden markov tree (HMT) respectively. Furthermore, the half tree increases the speed of HMT, and HMC can accurately initialize the parameters of the root nodes. The algorithm fuses MMSE despeckling technique, and by this method the noise-free wavelet coefficients are also estimated finely.更多还原