Daubechies

小波变换、EMD以及EMD结合小波变换几种常用的时频去噪方法的对比分析(Wavelet transform, EMD and the EMD of several commonly used wavelet transform time-frequency denoising Comparative Analysis)

说明：  基于小波包分解的脑电信号特征提取.格式是caj后缀的，大家看的时候可能得有CAJViewer 这个软件(Based on wavelet packet decomposition of the EEG feature extraction. CAJ suffix format is, we see there may be a time when the software CAJViewer)

双树复数小波变换程序，内附详细的使用说明，有小波变换函数可以选择(Dual-Tree Complex Wavelet Transform process, containing detailed instructions, you can choose a wavelet transform function)

本程序有去除脉搏波基线漂移、工频干扰和肌电干扰等功能(pulse wave )

小波变换的模极大值重建信号源程序 非常具有参考价值(wavelet transform modulus maxima reconstruction signal source is a very valuable reference)

基于morlet小波的小波熵的源代码，希望有用，共分享(the code of waveletentropy)

通过设计Visual C程序源码对简单易懂的一维信号在加上了高斯白噪声之后进行Daubechies小波、Morlet小波与Haar小波变换，从而的到小波分解系数；再通过改变分解的到的各层高频系数数进行信号的小波重构达到消噪噪的目的。在这一程序源码实现的过程中能直观地理解信号小波分解重构的过程与在信号消噪中的重要作用，和在对各层高频系数进行权重处理时系数的选取对信号消噪效果的影响。 可直接 (Design Visual C program source code on a simple one-dimensional signal with a Gaussian white noise Daubechies wavelet, Morlet wavelet and Haar wavelet transform, and thus to the wavelet coefficients and then change the decomposition to each storey wavelet reconstruction frequency coefficient of the number of signal to noise canceling noise. In the process of realization of this program source code can be intuitively understood the signal wavelet decomposition and reconstruction process and an important role in signal denoising, and the selection coefficient in the high frequency coefficients of the layers of the weight of processing noise cancellation signal impact. Can be directly)

说明：  真正的3 d双树离散小波变换(DWT)的一个数据集x是使用四个严格采样实现可分3 d DWTs在平行。然后subbands四DWTs结合适当。 (The real the 3 d Doubletree discrete wavelet transform (DWT) data set x four to strict sampling separable 3 d the DWTs In parallel. The then the subbands four DWTs combined appropriately.)

说明：  用于求取小波包变换后小波系数的熵值，对初学者很有帮助(Used to obtain the wavelet coefficients after wavelet packet entropy, helpful for beginners)

5*5的中值旋滤波，以前在这下的旋滤波发现有点问题，改了一下(5* 5 median spin filtering, spin under the previous filter in this issue found that a little bit, changed a bit)