二维经验模态分解域的新型HMT模型图像去噪  

Image denoising using BEMD domain based on a new HMT model

作　　者：吴昌健[1] 

机构地区：[1]辽宁师范大学计算机与信息技术学院,辽宁大连116029

出　　处：《微型机与应用》2015年第15期89-91,94,共4页Microcomputer & Its Applications

摘　　要：二维经验模态分解(Bidimensional Empirical Mode Decompositio,BEMD)是一种优秀的多尺度几何分析工具,特别适用于非线性、非平稳信号的分析处理。以BEMD与新型隐马尔可夫树(Hidden Markov Tree,HMT)模型理论为基础,提出了一种基于BEMD的新型HMT模型的图像去噪算法。该算法的基本思想是,首先对含噪图像进行BEMD变换,然后采用新型HMT模型对BEMD系数进行建模,并通过期望最大(EM)算法对图像BEMD的HMT模型参数进行估计,最后对训练后的BEMD系数进行逆变换,以获得去噪图像。仿真实验结果表明,该算法不仅拥有较强的抑制噪声能力,而且具有较好的边缘保护能力,其整体性能优于现有HMT图像去噪方案。BEMD transform is a kind of excellent multiresolution analysis tool, especially suitable for nonlinear and non-stationary signal analysis and processing. Based on BEMD transform and new Hidden Markov Tree（HMT） model, a new image denoising using BEMD domain new HMT models is proposed. Firstly, the BEMD transform is performed on the noisy image. Then,the BEMD coefficients are modeled using new HMT model, and the HMT model parameters are estimated utilizing maximum posterior probability. Finally, the trained BEMD coefficients are transformed back into the original domain to get the denoised image.Extensive experimental results demonstrate that the proposed method can obtain better performances in terms of both subjective and objective evaluations than those state-of-the-art denoising techniques. Especially, the method can preserve edges very well while removing noise.

关 键 词：图像去噪 二维经验模态分解 隐马尔可夫树 参数估计 

分 类 号：TN911.73[电子电信—通信与信息系统]