摘要
在小波阈值去噪方法的基础上,提出一种块匹配及小波变换技术相结合的图像去噪法,首先估算含噪图像的噪声方差,然后对图像进行分块匹配,构造各相似块的三维数据组,对其进行3D小波变换,再以噪声方差迭代形式,获得最佳软硬阈值对高低频系数分别做去噪处理,最后对低频DC系数做细节锐化运算。仿真结果表明,本算法既能有效地减轻图像中噪声,又具有较好的形状和细节保持能力,在图像的信噪比和主观视觉上都优于传统的软阈值、硬阈值、中、均值滤波去噪法。
Based on the multi-analysis wavelet threshold denoising method, in order to simuhaneously sharpen image details and attenuate noise, a blockmatching and wavelet transform filtering denoising approach is proposed. First, we estimate the noise variance of image, then compute the matching blocks, construct the 3D data array of those similar blocks, the high and low frequency sub-bands de-noised by the best soft threshold, hard threshold that result from the iterative calculation of noise variance respectively, at last, sharpen image details using DC coefficients of LL. Simulation results show that the algorithm can preserve and sharpen fine image details and effectively attenuate noise, at the same time, it has better performance than the traditional soft threshold, hard threshold, mean and average denoising methods.
出处
《四川理工学院学报（自然科学版）》
CAS
2009年第3期83-86,共4页
Journal of Sichuan University of Science & Engineering(Natural Science Edition)
基金
四川理工学院自然科学基金资助项目(2007ZR003)
关键词
图像去噪法
噪声方差
块匹配
小波变换
图像锐化
image denoising
noise variance
blockmatching
wavelet transform
sharpen image
