摘要
文章提出了一种基于小波域伪二维隐Markov树(P2DHMT)的图像的滤波新方法。首先建立了小波域的伪2DHMT模型,给出了基于EM、Baum-Welch等算法的模型参数估计方法;其次提出了一种基于最大后验概率准则的P2DHMT最优图像滤波算法;最后给出了图像去噪算法的实现过程。实验结果表明该方法可以在保存图像细节特征的情况下有效地抑制图像的噪声。
A novel image filter approach is proposed based on Pseudo 2D Hidden Markov Model in wavelet domain. First,Pseudo 2D Hidden Markov Tree（P2DHMT） model in wavelet domain is established.The parameters can be estimated by EM algofithm,Baum-Welch algorithm.Second,optimal filtering,algorithms for hidden Markov model.is presented employing maximum a posteriori criterion.The wavelet coefficient of image can be estimated by the filter algorithm,Appling this method to simulated image and real SAR image,the result shows that this method can effectively reduce noise,while keeping image detail.
出处
《计算机工程与应用》
CSCD
北大核心
2006年第4期45-47,83,共4页
Computer Engineering and Applications
基金
国家自然科学基金资助项目(编号:10274060)
