try this one prepared by the neural network training and testing of the source c...

快速提升小波变换的源程序，可以处理一维和二维信号和图像。-quickly upgrade wavelet transform of the source, can handle one-dimensional and two-dimensional signals and images.

lifting scheme指南-lifting scheme Guide

利用小波变换对信号进行分析处理,现代信号处理,胡广书编-using wavelet analysis of signal processing, modern signal processing, Mr Kwong Textbook

小波变换的visual c++实现源代码，有效实现了小波的快速变换，得到很好的恢复图像，是个很有价值的源程序。-wavelet transform visual c achieve source code, the effective realization of the fast wavelet transform, well restored images, is a valuable source.

it is a function of wavelet transform

一个小波边缘检测的MATLAB程序,一个小波边缘检测的MATLAB程序-One of the MATLAB wavelet edge detection procedure, one of the MATLAB wavelet edge detection procedure

采用5/3小波提升算法编写的一维和二维信号小波分解与重构程序。给出了图象分解的演示示例。-using 5/3 wavelet algorithm to enhance the one-and two-dimensional wavelet decomposition and reconstruction process. Given the decomposition of images demonstration examples.

图像正交变换，包括傅里叶变换，离散余弦变换，沃尔什变换，小波变换等-Orthogonal transformation images, including Fourier transform, discrete cosine transform, Walsh transform, wavelet transform, etc.

小波变换源代码，实现了几种不同的小波变换-Wavelet Transform source code, to achieve a number of different wavelet transform

关于二维小波变换的程序 [精华] 说明：此算法重在概念，速度并不是很快。因为FOR循环的缘故。此程序从循环矩阵的观点出发，把圆周卷积和快速幅里叶变换建立了联系。实现了分解和无失真重构。它只做了一层分解，即将256x256图形分解成为64x64的四个图形，避免了使用WKEEP()的困惑。主要思想为用小波滤波器族构造正交阵W,变换写为B=W*A*W ,反变换为：A=W *A*W,这与所有正交变换无异。W为循环正交矩阵，因此可用FFT实现快速运算，难点就在重构矩阵上。若用矩阵概念明确，一个共扼转置可以搞顶。但FFT的使用必须找到与分解序列的关系。-on Wavelet Transform procedures [best] Note : This algorithm important concept is not speed quickly. Because FOR cycle nowadays. This program cycle from the point of view of Matrix, circular convolution and fast Fourier transform is established links. To achieve the decomposition and reconstruction without distortion. It only had one decomposition, 256x256 graphics will be 64x64 decomposition of the four graphics, avoiding the use of WKEEP () confusion. The main idea of using wavelet filter generator orthogonal array W, write to transform B = W* A* W, anti-Transform : A = W* A* W, which is orthogonal transformation all the same. W cycle orthogonal matrix, can be used to achieve rapid FFT computation, on the dif