基于小波域隐马尔可夫混合模型的SAR图像降斑算法

在小波域马尔可夫随机场(MRF)和隐马尔可夫树(HMT)的基础上,提出了一种新的合成孔径雷达(SAR)图像降斑算法.该算法在对乘性噪声不取对数变换的情况下,融合了贝叶斯最小均方误差(MMSE)抑制噪声技术.为了提高HMT的速度,采用了一个新的隐马尔可夫半树模型,该模型考虑了小波系数的持续性和聚类性,分别用HMT和MRF刻画.仿真结果表明该算法在抑制斑点噪声的同时,有效的保持了边缘,避免对数变换带来的一些误差,取得了好的效果,其速度比HMT模型提高了二十倍.

SAR images despeckling based on wavelet and hidden Markov mixture model

An efficient despeckling method was proposed based on hidden Markov random field (MRF) and hidden Markov tree (HMT) in the wavelet transform for synthetic aperture radar images. The algorithm fused minimum mean square error despeckling technique in the circumstance of avoiding the log-transform. In order to increase the speed of HMT, a new hidden Markov half tree model was introduced. The clustering and the persistence of wavelet coefficients were taken into account, which were characterized by MRF and HMT respectively. Experimental results show that the method presented achieves good performance in terms of noise suppressing and edges preserving, and the running time is less than HMT by about twenty.