结合Contourlet变换的Bayes自适应图像去噪算法  

Adaptive image denoising algorithm integrating bayes with contourlet transform

作　　者：丁亮[1,2] 汪烈军[1] 贾振红[1] 

机构地区：[1]新疆大学信息科学与工程学院 [2]69018部队

出　　处：《激光杂志》2013年第6期43-45,共3页Laser Journal

基　　金：国家自然科学基金项目(项目编号:61261036)

摘　　要：提出了结合Contourlet变换的Bayes自适应图像去噪算法。充分利用Contourlet变换的局部性、多分辨率、各向异性等优点。通过Contourlet变换得到图像不同尺度不同方向上Contourlet系数矩阵,实现了建立在对图像多尺度几何分析基础上Bayes估计自适应图像去噪算法。实验表明,新算法能够获得良好的视觉效果并且有效地提高了去噪图像的PSNR值,同时有效的避免了"过扼杀"系数现象,更好地保留了图像的纹理和细节。An adaptive image denoising algorithm integrating Bayes with Contourlet transform was proposed, which utilizing the Contourlet transform＇s advantages of multi - resolution and anisotropy. Different scale matrixes of Contourlet coefficient of the image can be acquired through Contourlet transform, therefore, the improved image denoising algorithm made it possible to realize the new adaptive Bayes threshold denoising based on multi - scale geometric analysis of the image. The experiment results have demonstrated that the proposed method can significantly improve the PSNR value, visual effect and avert the problem of snuffing out the coefficients effectively while preserving image textures and details.

关 键 词：图像去噪算法 CONTOURLET变换 多尺度几何分析 Bayes阈值 

分 类 号：TN248.1[电子电信—物理电子学]