wavetest0

利用全局去噪和分层去噪实现信号的小波去噪，实现成功(Implementation of wavelet denoising, global denoising and level-dependent denoising)

本文研究的是利用奇异性指数实现对设备状态监视和故障预警功能。首先用LabVIEW 采集传感器电信号，然后对采集到的电信 号进行小波变换得到信号包络奇异点的奇异性指数，即设备运行状态的特征值(In this research project, we use LabVIEW to do wavelet-transformation to the signal collected by sensors in order to get singularity index from the singular points of the particular signal’s envelope. Through comparing this eigenvalue to the figures in a standard database, we can know the functional status of the instruments. Not only the diagnoses are continuously displayed, but alarms about signs of dysfunction will also be raised in a proper way. In other words, both the supervision of the instrument and warning of dysfunction can be achieved)

运行emd,得到输入信号X的固有模态函数（imf），每列代表着信号X的一阶分量，X就是我们采集的信号。 hhtdn是hht去噪的程序，可以不看，仅看emd 和 findpeaks wvtdn是wavelet去噪程序，如果只是分解信号，仅用下面的语句即可 [C,L] = wavedec(x_noise,dl,wn); x_noise就是我们采集的信号，dl是分解的阶数，wn是分解的母小波名称(% Empiricial Mode Decomposition (Hilbert-Huang Transform) % imf = emd(x) % Func : findpeaks)

利用小波变换实现表面肌电信号消噪的程序，还包括一些对比分析。(Wavelet transform SEMG denoising procedures, including some comparative analysis.)

我实现的离散小波变换，可直接运行，采用三级变换(Discrete wavelet thransform)

说明：  可导入comtrade格式录波文件，进行谐波分析(It can import COMTRADE format recording file for harmonic analysis)

使用DCT和小波变换的数字音频水印生成/解码程序，可存活于方差0.05的高斯白噪声及不同程度的剪切。内含程序/测试文件介绍。(Digital wartermarking encoding/decoding program with DCT and wavelet transform. nice robustness against Gaussian noise with variance less than 0.05 and any type of cropping attack. functions/test files introduction included)

小波包分解包括重构信号问题程序，这是一个matlab程序，经过测试可用，数据采用自己的数据。(Including the reconstructed signal wavelet packet decomposition problem program, which is a matlab program, tested available data using their own data.)

结合离散小波变换(DWT)技术，利用DWT与HVS相结合的良好特性，探索和建立了一种新的图像质量客观评价方法(taking advantage of the similarities of discrete wavelet transform (DWT) with HVS，a new methods for integrated image quality assessment)

设计陷波滤波器，建立原始信号，对其采样，将采样信号通过滤波器进行滤波(Design trapped wave filter, establish the original signal, the sampling, will sampling signal through the filter to filter)