基于第二代Bandelet变换的自适应阈值去噪算法  

An Adaptive Threshold De-noising Algorithm Based on Second Generation Bandelet Transform

作　　者：杨居义[1] 

机构地区：[1]绵阳职业技术学院计算机科学系,四川绵阳621000

出　　处：《微电子学与计算机》2009年第8期219-222,共4页Microelectronics & Computer

摘　　要：针对目前最新发展的Bandelet变换的特点,文中提出了基于第二代Bandelet变换的自适应阈值去噪算法,充分利用图像内在的几何正则性,自适应获得图像的最优表示,通过噪声强度设置阈值,并利用软阈值函数实现图像去噪.实验结果表明,该算法比小波算法去噪后的图像在视觉效果及客观指标上都有显著提高.Aiming at the latest development d Bandelet transform characteristic, an adaptive threshold de-noising algorithm based on second generation Bandelet transform is mentioned in this paper. It takes the full advantage of geometry regularity in image, gets the best expression of the image, sets up the threshold by noise intensity, and realizes de-noising in image under the use of soft-threshold function. The experiment result shows that the image processed by this algorithm can get better efforts than wavelet algorithm both in visual and objective quality targets.

关 键 词：第二代BANDELET变换 自适应阚值去噪 小波变换 图像 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]