【摘要】 介绍了相关去噪和模极大值去噪两种传统的去噪方法的原理,并针对它们在滤除噪声上的明显不足,提出了一种基于区域相关的滤波新算法。它克服了通常相关算法中由于各尺度间小波系数的偏移引起的判断准确率低的缺点,进而对小波域滤波之后的系数进行模极大值处理,以去除残留的噪声系数。这种算法把两种思想截然不同的小波滤波方法合在一起,并作了一些改进,充分考虑了信号相邻系数间的相关性,较好地解决了保护信号局部特征与抑制噪声之间的矛盾。仿真结果验证了算法的有效性。更多还原

【Abstract】 The paper introduces the two great-traditional denoising methods which are correlation denoising and max-modular denoising.And then aimed at the shortcoming of these two methods,it presents a new filtering algorithm based on the region relativity of the wavelet coefficients,which overcomes the disadvantage of the available relativity-based algorithm which is inaccurate in computing the relativities of the deflected wavelet coefficients.When the filtered coefficients in the wavelet domain are obtained,a max-modular method is used to get rid of the noisy coefficients among them.The algorithm presented in this paper is the combination and the improvement of two different wavelet filtering methods.And this method can settle the conflict between the signal’s local characteristic protection and noises restriction commendably.Finally,the validity of the proposed method is verified by the results.更多还原