自适应阈值的小波去噪改进算法研究

基于小波变换的图像去噪方法在消除噪声的同时,可有效保留图像边缘细节信息,是近阶段图像去噪领域研究与应用的热点.现有的基于小波阈值法的去噪算法多为全局阈值,易引起边缘模糊.因此,在阐述小波去噪基本原理的基础上,将小波变换和多尺度边缘检测两者结合,充分考虑小波分解不同层数的特性,提出一种具有自适应阈值的小波图像去噪改进算法.实验表明,改进算法与传统去噪方法(维纳滤波法)及一般小波阈值法(VisuShrink阂值法、NormalShrink阈值法、BayesShrink阈值法)相比,可有效去除多种程度的加性高斯白噪声,更好保留图像边缘细节信息.

Improved alorithm with auto-adaptive threshold for wavelet image denoising

Image denoising based on wavelet transform can effectively keep the details of image edge,and become the hotspot in research and application of image denoising. Now the denoising algorithm based on wavelet threshold method is mostly the global threshold, which can easily cause the edge blur. In this paper, on the basis of the basic principle of wavelet denoising and giving full consideration to the decomposition level of wavelet characteristics, an adaptive threshold of wavelet image denoising algorithm which uses the wavelet transform and multi-scale edge detection is proposed. Experimental results show that, compared with traditional algorighm(the Wiener filtering) and general wavelet thresholding(VisuShrink Thresholding, NormalShrink Thresholding, BayesShrink Thresholding), the improved algorithm can effectively remove various degrees of additive white gaussian noise, keep the details of image edge better.