摘要
该文提出了一种基于小波域统计混合模型的图像降噪方法。该方法首先利用尺度间模型,将小波系数分成两类:有意义系数和无意义系数;然后在小波域同层模型中运用最大后验概率估计方法,从有意义系数中恢复出原始系数。文章给出了算法的完整步骤。实验结果及分析表明了该方法的有效性。
In this paper, a novel image denoising method based on statistical mixture model in wavelet domain is proposed. Firstly, the wavelet coefficients are classified as significant and insignificant coefficients by using interscale statistical model. Secondly, Maximum A Posteriori （MAP） estimator based on intrascale statistical model is used to restore the noisy wavelet image coefficients. A completive algorithm is presented to implement this idea. Experimental results and analysis are given to demonstrate the validity and effectiveness of the proposed method.
出处
《电子与信息学报》
EI
CSCD
北大核心
2005年第11期1722-1725,共4页
Journal of Electronics & Information Technology
关键词
小波变换
最大后验概率
混合模型
图像降噪
Wavelet transform, Maximum A Posteriori （MAP）, Mixture model, Image denoising
