code

混合高斯模型来进行前景背景区分的matlab实现代码(Gaussian mixture model to distinguish between foreground and background matlab implementation code)

haar二维小波分解，输出四个分解系数子矩阵.将系数重构，得到最终的输出矩阵(haar two dimension wavelet decompositon)

最终的权值矩阵就是滤波器的系数，处理信号的时频分析，阵列信号处理的高分辨率估计，包括最小二乘法、SVM、神经网络、1_k近邻法,有较好的参考价值，结合PCA的尺度不变特征变换(SIFT)算法，计算加权加速度,旋转机械二维全息谱计算。( The final weight matrix is ??the filter coefficient, When processing a signal frequency analysis, High-resolution array signal processing estimates, Including the least squares method, the SVM, neural networks, 1 _k neighbor method, There are good reference value, Combined with PCA scale invariant feature transform (SIFT) algorithm, Weighted acceleration, Rotating machinery 2-d holographic spectrum calculation.)

The program get a spectrum with eigenvector method

一个关于广义最小二乘法的matlab语言程序和仿真框图(A generalized least squares matlab language programming and simulation block diagram)

Este arquivo representa um prova de dispositivos de engenharia elétrica

一个维纳滤波算法,利用MATLAB编写的,解压后可直接调用(a Wiener filter algorithm, using MATLAB, unpacked directly call)

frank-wolfe的MATLAB实现(an example for frank-wolfe by matlab.)

这个小程序可以简单的选取回归分析中box-cox变换的lambda值，通过matlab强大的矩阵运算功能可以通过简单比较RSS值的方式来得到lambda，使用时只需要输入实验值矩阵X和Y即可，然后输入minRSS（X，Y）就能简单明了的看到最优变换参数 （这个程序虽然小但是学习回归分析的初学者学习时都会经过这个算法，上载时因为不到1KB所以我又放了个文本格式的）(This little program can select a simple regression analysis of box-cox transformation of lambda values, through the powerful matlab function matrix calculation can be relatively simple way of RSS to get the value of lambda, the use of only need to enter the experimental data matrix X and Y can be , and then enter the minRSS (X, Y) will be able to see the simple optimal transformation parameters)

说明：  随机共振的一些论文，很有权威，对研究非线性双稳随机共振系统的人很有帮助(Some of the papers stochastic resonance, very authoritative, the study of nonlinear bistable stochastic resonance systems were very helpful)