摘要
为了在滤除图像噪声的同时又能保留图像边缘,提出了一种新的基于小波变换的图像滤波法。利用小波变换对噪声图像进行多分辨率分解,得到对应的低频近似信号分量和高频细节信号分量。提取低频系数和高频系数,分别对低频系数进行增强,对高频系数进行锐化,再对锐化后的高频系数进行阈值处理,获得噪声系数,利用锐化前的高频系数减去处理后的高频系数,得到最终的高频系数并通过小波重构得到去噪图像。
A new method of image filter for preserving the edge as well as reducing the noise of images based on the wavelet transformation is put forward. The image with noise is decomposed into multi-resolution levels and the corresponding detail and approximation coefficients are obtained. The low frequency coefficients and the high frequency coefficients are extracted, so as to enhance the low frequency coefficients and sharpen the high frequency coefficients separately. The high frequency coefficients are sharpened in dealing with the threshold value, and the noise coefficients are obtained. The final high frequency coefficients are then derived by subtracting the original value with the present value, and the noise-reduced image can be reconstructed through the inverse wavelet transformation.
出处
《江苏科技大学学报(自然科学版)》
CAS
北大核心
2007年第4期46-49,共4页
Journal of Jiangsu University of Science and Technology:Natural Science Edition
关键词
图像滤波
小波变换
边缘保留
小波系数
锐化
image filter
wavelet transform
edge preserving
wavelet coefficients
sharpening