针对医学图像的小波阈值去噪方法  

The Method of Wavelet Threshold Denoising for Medical Image

作　　者：张艳[1] 程小平[1] 

机构地区：[1]西南大学计算机与信息科学学院,重庆400715

出　　处：《西南师范大学学报（自然科学版）》2011年第3期245-249,共5页Journal of Southwest China Normal University(Natural Science Edition)

摘　　要：为了改善小波阈值去噪算法中硬阈值和软阈值存在的不足,提出一种新的小波阈值去噪方法.该算法在进行小波阈值去噪前,先将图像分割成背景平坦区域和细节区域两部分,然后分别进行小波阈值去噪,最后融合两图像从而获得去噪图像.在分别进行小波阈值去噪时,利用迭代法进行阈值选择,采用"软、硬阈值折中"阈值函数.根据对医学图像去噪的仿真实验结果表明,该算法在去噪效果上均优于传统的软硬阈值方法.In order to improve some shortcomings in the hard threshold and soft threshold of wavelet threshold denoising,the authors propose a new wavelet threshold denoising method.Before wavelet threshold denoising,segment image into two regions：background and detail,and then wavelet threshold denoising respectively,and finally gets the fusion of the two images.In separate wavelet threshold denoising using the iterative threshold selection method,using ＇soft and hard threshold compromise＇ threshold function.Based on the show of simulation results of medical image denoising,this algorithm in PSNR and MSE is superior to the traditional hard and soft threshold method.

关 键 词：小波变换 阈值去噪 阈值函数 医学图像 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]