基于多小波的局部背景隐马尔可夫模型图像去噪

小波域局部背景隐马尔可夫模型（LCHMM）可获得尺度内的相关性和局部的统计特征，并且复杂度小，多小波分析在图像去噪方面有很好的性能。利用多小波分析和局部背景隐马尔可夫模型各自在图像去噪方面的优势，将两者结合起来，提出了一种基于多小波的局部背景隐马尔可夫模型（M—LCHMM）图像去噪算法。算法主要有两步：局部背景隐马尔可夫模型去噪处理和均值处理。该算法简单有效，仿真试验表明M—LCHMM的去噪效果优于目前许多已有的去噪算法。

Image Denoising Based on Multiple Wavelet and Local Contextual Hidden Markov Model

Wavelet-domain local contextual hidden Markov model （LCHMM） can exploit both the local statistics and the scale of relevance in a low computational complexity. Multiple wavelet representations have excellent performance in image denoising. In this paper, combining the multiple wavelet representations with the LCHMM and using their advantages in image denoising, a new algorithm is proposed, which is called M- LCHMM. This algorithm, which is simple and effective, has tWO major steps： local contextual hidden Markov model denoising of the wavelet coefficients and an averaging of the denoised images. Simulation results show that the proposed M-LCHMM can achieve the state-of-the-art image denoising performance in a low computational complexity.