一种基于SSNF的新型阈值滤波去噪算法

介绍了一种空间选择性的噪声滤波(SSNF)方法,并在此工作之上提出了一种新的自适应于小波变换尺度的阈值函数,从而对经SSNF滤波之后的小波系数进行了进一步的阈值处理,以去除残留在系数中的噪声部分。仿真试验和理论分析表明,相比其它传统的去噪方法,该方法的优点在于:所得到的小波系数不仅连续性好,而且更加接近于未加噪信号的小波系数,阈值函数具有很大的灵活性和自适应性,并适用于一些掺杂非白噪声的场合。

A new kind of threshold filtering de-noising algorithm based on SSNF

First,the spatially selective noise filtration(SSNF) method was introduced,and based on this work a new threshold function self-adaptive to the wavelet transform scale was presented,and then the wavelet coefficients gained from SSNF were through further threshold disposal,by this way the remnant noise in the coefficients was wiped off.By the simulation and theoretical analysis we conclude that the proposed method has some advantages compared with those traditional de-noising me-thods.These advantages include: the gained wavelet coefficients not only have good continuity,but are closer to the original signal's wavelet coefficients;the threshold function has large flexibility and self-adaptivity.This method is also applicable to some situations in which the original signals are mixed by the non-white noises.