摘要
小波系数在一个小邻域里具有相似性,若充分利用周围小波系数的信息,分块对小波系数进行阈值去噪操作,可以在保持图像平滑的同时,尽量多地保留图像的细节信息。对Cai&Silverman的方法进行分析,将其拓展到二维信号去噪,并提出一种基于小波块阈值数字图像去噪方法。实验结果表明,该方法不但能保留更多的图像细节,而且具有较好的峰值信噪比。
Wavelet coefficients are correlated in a small neighbourhood.We can take into account ncighbour wavelet coefficient, and threshold the wavelet coefficients.Thus can produces smoother and clearer denoised images.In this paper,we discuss the Cai＆Silverman＇s idea and extend it to the 2D case,then we present a de-noising method based on 2D wavelet thresholds. The experimental results show that the proposed method can keep images＇ detail information and get higher PSNR for the denoised
出处
《湖南理工学院学报（自然科学版）》
CAS
2006年第1期31-33,39,共4页
Journal of Hunan Institute of Science and Technology(Natural Sciences)
关键词
小波变换
二维小渡阈值去噪
相邻小波系数
数字图像
wavelet transform
2D wavelet threshold de-noising
ncighbouring wavelet coeffieients
digital image
