摘要
提出一种基于ROLD统计量的混合噪音线性滤波算法(RLMF)。算法把用来检测脉冲噪音的ROLD统计量运用于混合噪音的滤波算法上,提高了混合噪音中脉冲噪音成分的检测效率,它不仅适用于恢复被混合噪音污染的数字图像,而且也适用于恢复被纯脉冲噪音或纯高斯噪音污染的数字图像。仿真实验证明,RLMF滤波后的图像视觉效果和PSNR均优于已知的同类滤波器。
A mixed noise linear filter based on ROLD statistics is proposed, RLMF for short.This algorithm applies the ROLD statistics for detecting the pure impulse noise to the mixed noise filter.Its efficiency of detecting impulse component is increased in mixed noise.It is not only suitable for digital images corrupted with mixed noises（impulse noise and Gauss- ian noise）,and suitable for digital images corrupted with pure impulse noise or pure Gaussian noise.The experiments show that,visual effects and PSNR of denoised image produced by RLMF are superior to the known similar filters.
出处
《计算机工程与应用》
CSCD
北大核心
2010年第31期175-177,191,共4页
Computer Engineering and Applications
基金
河南省教育厅自然科学基金(No.2009C110004)
关键词
数字图像
ROLD统计量
混合噪音
滤波算法
digital image
Rank Ordered Logarithmic Difference（ROLD） statistic
mixed noise
filter algorithm
