wavelet-package

利用小波变换，通过设置阈值对图像进行去噪。 首先对图像用小波进行2层分解，然后通过设置硬阈值和软阈值的方法，对图像高频部分去噪。(The wavelet transform is used to denoise the image by setting threshold. Firstly, the image is decomposed into 2 layers by wavelet, and then the high frequency part of the image is denoised by setting the hard threshold and the soft threshold.)

single level ,two level, multi level dwt for the denoising

对信号f(x)进行三层小波变换，得到小波系数d1,d2,d3和逼近系数a3，我想求出各个频率段的能量（能量=小波系数的平方和），很不错的应用程序(The signal f (x) for three wavelet transform, wavelet coefficients d1, d2, d3 and approximation coefficients a3, I would like to find the energy of each frequency band (energy = sum of squared wavelet coefficients), very good application)

对含噪声信号进行小波变换；对变换得到的小波系数进行某种处理，以去除其中包含的噪声；对处理后的小波系数进行小波逆变换，得到去噪后的信号(Containing noise signal wavelet transform the wavelet transform coefficients to a treatment to remove the noise contained therein the treated wavelet inverse wavelet transform coefficients to obtain a denoised signal)

这是一本关于高维(三维以上)小波的书，对做三维的朋友应该有参考价值，所以特意贡献出(This is one of the high-dimensional (3-D above) wavelet, the book right to do 3D friends should have reference value, So deliberately contributing)

对信号进行局部均值分解，得到一系列具有物理意义的pf分量(Local mean decomposition of the signal to obtain a series of having a physical meaning PF component)

一种基于改进谱减法的语音去噪新方法，对于想学习谱减法的同学有很大的帮助(A kind of based on improved spectral subtraction speech denoising, a new method to want to learn the spectral subtraction classmates have a lot of help )

脉搏信号分析 （1）设计滤波器，实现对脉搏信号的噪声抑制和基线纠漂； （2）时域分析：波形特征检测； （3）功率谱分析：对消噪后的信号进行功率谱分析。要求计算信号的功率谱，功率谱峰值，峰值频率。 (Pulse signal analysis (1) design the filter, to achieve the pulse signal noise suppression and baseline drift correction (2) time-domain analysis: waveform feature detection (3) power spectral analysis: noise cancellation signal after the power spectrum analysis . Require calculation of the signal power spectrum, peak power spectrum, peak frequency.)

利用CVX进行优化，得到期望加权矢量，从而的得到需要的波束形成(CVX using optimized to give the desired beamforming)

Multiscale Curvature-Based Shape Representation Using B-Spline Wavelets