ZX_TFPF

小波域图像滤波，采用双变量收缩函数发实现(Bivariate Shrinkage Functions for Wavelet-Based Denoising)

对于非平稳信号，小波多尺度分解师一种有效的信号去噪方法，在小波去噪方法的基础上，提出了基于Lipschitz指数的小波阈值去噪方法(For non-stationary signals, signal de-noising of wavelet multi-scale decomposition ' is an effective method, wavelet denoising method based on proposed wavelet thresholding denoising method based on the Lipschitz exponent)

在理解了离散小波变换的基本原理和算法的基础上，通过设计VC程序对简单的一维信 号在加上了高斯白噪声之后进行Daubechies小波、Morlet小波和Haar小波变换，从而得到小波分解系数；再通过改变分解得到的各层高频系数进行信号的小波重构达到消噪的目的。在这一程序实现的过程中能直观地理解信号小波分解重构的过程和在信号消噪中的重要作用，以及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。(In understanding the basis of discrete wavelet transform basic principles and algorithms, through the design VC program Daubechies wavelet, Morlet wavelet and Haar wavelet transform after a simple one-dimensional signal plus a Gaussian white noise, resulting wavelet coefficients wavelet reconstruction signal again by changing the layers to achieve high-frequency coefficients decomposed noise cancellation purposes. Can intuitively understand wavelet decomposition and reconstruction process and an important role in signal de-noising in the process of implementation of this program, and in the high-frequency coefficients when the weight of the layers selected for signal processing coefficients denoising affected.)

小波变换的程序包括小波变化 小波分解小波包分解程序(Wavelet transform procedures include wavelet transform wavelet decomposition wavelet packet decomposition program)

全套voltera滤波器检测目标回波信号，且全部都是m文件，没有加密，并附原始信号，以供验证。(Complete voltera filter detection of the target echo signal, and all the m files are not encrypted, together with the original signal, for verification.)

小波变换用于半导体激光器可靠性分析的全部程序。我只完成了其中一部分，请斑竹验收(wavelet transform for the semiconductor laser reliability analysis of all procedures. I only completed part, illustrated acceptance)

小波神经网络,实现曲线拟和，数据分类，故障诊断，学习速度快于传统BP。(Wavelet neural network, and realize curve fitting, data classification, fault diagnosis, learning faster than the traditional BP.)

用小波方法把脑电信号进行频段分解，并画出图形(Using wavelet method to decompose the eeg signals into frequency bands, and draw the graph)

编写MATLAB程序，产生协方差函数为C(τ)=9??^(?10|??| )的零均值平稳高斯过程，产生一条样本函数.测量所产生样本的时间自相关函数，将结果与理论值比较。(Procedures for the preparation of MATLAB produced C covariance function (tau) =9^ (- 10||) zero mean stationary Gauss process to produce a sample function. Measuring the resulting sample time autocorrelation function, the results will be compared with the theoretical value.)

根据采集的时域数据，做fft得到其频谱，并计算其功率谱密度(According to the collected time-domain data, the frequency spectrum is obtained by FFT and the power spectral density is calculated.)