贝叶斯正则化算法提高 BP 网络的推广能力

FFT傅里叶快速变换，把变换的结果放到文件中，并用matlab绘制图形。( FFT the Fourier fast transformation, puts the transformation result to the document, and draws up the graph with matlab. )

阵列天线设计，提供方向图示意，计算阻抗等，很适用于初学者(array antenna design)

本程序介绍了一种PPB滤波算法来去除图像中的加性高斯噪声，09年才提出来的新算法非常的高效，去噪效果相当好(This procedure describes a filtering algorithm to remove the image PPB in additive Gaussian noise, 09 years to put forward a new algorithm is very efficient, very good denoising effect)

i have wote a matlab code to track human in vidoe and this is well runing but i need a code for huma detection in maltab if anyone have thenm plz mail me on my adrees

A律十三折线量化编码的实现。 (A thirteen line Quantization law implementation.)

Matlab Engineering Analysis

很全面的matlab优化算法的代码集合，包括蚁群免疫等算法都在，很细致。(Very comprehensive collection of matlab code optimization algorithm, including the immune ant colony algorithms such are very detailed.)

means3vm算法，matlab 是实现半监督学习的较好的方法，能够对多种数据集进行测试，代码中包含例子，下载即可以使用(means3vm algorithm, matlab is better to achieve semi-supervised learning methods can be tested on a variety of data sets, the code contains examples that can be used to download)

生成调制信号数据集 信号集包括2ASK，2PSK，2FSK，4ASK，4PSK，4FSK，8ASK，8PSK，8FSK，16QAM，64QAM，信号比范围-20db至18dB，调制到中频。 信号参数 符号率fb：1e3 载波频率fc：2e3 采样频率fs：64e3 信号频道：awgn，rayleigh(Generating Modulation Signal Data Set Signal sets include 2ASK, 2PSK, 2FSK, 4ASK, 4PSK, 4FSK, 8ASK, 8PSK, 8FSK, 16QAM, 64QAM, signal ratio range - 20 dB to 18dB, modulated to intermediate frequency.)

图像配准常常是作为其他图像处理应用的前处理步骤使用的，往往用于图像的对准、目标识别与定位。介 绍了在MATLAB系统中，如何应用MATLAB的IPT函数对图像进行配准的方法：首先用人机交互的方法在待配 准图像与基准图像之间进行图像配准所必需的匹配控制点的选取，然后用这些匹配控制点来计算两图像的某种空 间几何变换关系，最后利用这个空间几何变换关系对待配准图像进行几何变换，获得配准结果。实验表明，该方 法实现容易，且非常有效。 (Image registration is often as other image processing applications, the use of pre-treatment steps are often used for image alignment, target identification and location. Introduced the system in MATLAB, how to use MATLAB function of the IPT image registration method: First of all, employing the method of cross-machine in question and the baseline image registration between image image registration by matching control points necessary for the selection, and then use these control points to match the two images to calculate the geometric transform the relationship between a certain space, and finally make use of the space relationship between the treatment of geometric transformations with quasi-geometric image transformation, the results obtained registration. Experiments show that the method easy and very effective.)