danchunxingfa

this is code of the ica algorithm

matlab入门级别的电子书，主要是基本的操作运算和一些matlab的常用函数汇总(matlab entry-level e-books, the basic operation of arithmetic and some of the commonly used matlab function summary)

主要是关于彩色图像边界提取的源代码。里面包含有待处理的图片以及所有的源代码，即M文件。(Mainly on the color image edge extraction of the source code. Which contains the picture to be addressed, as well as all the source code, that is, M documents.)

投影寻踪等级评级模型求解和遗传算法的实现(Projection Pursuit Model and Genetic Algorithm for Solving)

Koch分形曲线计算地图边界，此程序能较好的计算任意凸图形边界长度。程序计算步骤为：提取图形边界，将边界点按与中心点所成角度排序，每隔一定角度取一边界点，对选取的边界点做3次迭代Koch分形(Koch fractal curve map boundary, this procedure can be better calculated arbitrary convex graphical boundary length. The program calculation steps: to extract graphical boundary, the boundary points and the center point of the angled Sort every angle to take a boundary point, done the 3 iterations Koch fractal selected border points)

实现聚类K均值算法： K均值算法：给定类的个数K，将n个对象分到K个类中去，使得类内对象之间的相似性最大，而类之间的相似性最小。 缺点：产生类的大小相差不会很大，对于脏数据很敏感。 改进的算法：k—medoids 方法。这儿选取一个对象叫做mediod来代替上面的中心 的作用，这样的一个medoid就标识了这个类。步骤： 1，任意选取K个对象作为medoids（O1,O2,…Oi…Ok）。 以下是循环的： 2，将余下的对象分到各个类中去（根据与medoid最相近的原则）； 3，对于每个类（Oi）中，顺序选取一个Or，计算用Or代替Oi后的消耗—E（Or）。选择E最小的那个Or来代替Oi。这样K个medoids就改变了，下面就再转到2。 4，这样循环直到K个medoids固定下来。 这种算法对于脏数据和异常数据不敏感，但计算量显然要比K均值要大，一般只适合小数据量。(achieving K-mean clustering algorithms : K-means algorithm : given the number of Class K, n will be assigned to target K to 000 category, making target category of the similarity between the largest category of the similarity between the smallest. Disadvantages : class size have no great difference for dirty data is very sensitive. Improved algorithms : k-medoids methods. Here a selection of objects called mediod to replace the center of the above, the logo on a medoid this category. Steps : 1, arbitrary selection of objects as K medoids (O1, O2, Ok ... ... Oi). Following is a cycle : 2, the remaining targets assigned to each category (in accordance with the closest medoid principle); 3, for each category (Oi), the order of selection of a Or, calculated Oi Or replace the consumption-E (Or))

提出了一种基于小波变换可抗击几何攻击算法。在水印嵌人 部分# 首先对宿主图像进行小波分解 、 嵌人水印, 再通过余弦变换求得该水印图像的一个特征向量, 利用该特征向量和水印 信息通过12 31 函数生成一个二值逻辑系列, 将该序列作为密钥/ 水印提取部分为# 对待测的图像进行小波变换, 再通过余 弦变换求得待测图像的一个特征向量, 利用该特征向量和密钥通过1 2 4 1 函数提取出水印。(Proposed based on wavelet transform algorithm can be against geometric attacks.# Watermark embedded in the first part of the host image wavelet decomposition, embedded watermark, and then through the cosine transform of the watermarked image obtained by a feature vector, using the feature vector and the watermark information through 1231 to generate a binary logic function series, the sequence as the key/watermark extraction part is# treated transform measured image, and then obtained through the cosine transform the image of a feature vector under test, using the feature vector and the key function by 1241 to extract the watermark .)

a good PDF file for introduction of ECG

说明：  是关于对角PCA算法的matlab代码，包含对角化代码和对角化PCA代码(Is the angle of PCA algorithm on matlab code, including code and diagonalization Diagonalization PCA code)

The Matrix Cookbook 这个资料包含了矩阵的各种运算公式，对于不熟悉矩阵运算的朋友可以看看(The Matrix Cookbook This information contains a variety of computing equation matrix matrix operations are not familiar friends can see)