高斯赛德尔 高斯赛德尔 高斯赛德尔 高斯赛德尔 高斯赛德尔

AI::Categorizer is a framework for automatic text categorization. It consists of a collection of Perl modules that implement common categorization tasks, and a set of defined relationships among those modules

应用背景主要是对手写字体进行识别，首先训练手写字体库，可在官网下载手写字体库。对样本库中的样本进行训练，进而识别多种手写字体，主要针对0--9这10个数字，可以准确进行识别。关键技术主要采用C和OPENCV混合编程，提取样本HOG特征，运用SVM分类器对样本进行分类识别。获得训练分类器后，可以自己手写一个0--9中的任意字体，放入分类器中进行识别。

采用滤波方法来求取地球大气风暴轴EKE的一种算法，国内外已经有专家进行过这方面的研究，这是个人自编的算法代码

矩阵的最大特征值的幂法. 对于工程计算而言，矩阵的特征值和特征向量都是相当重要和常见的数据，这里给出的幂法是一种常见的求解方法，用的是迭代的思想。 符号说明: 1A为待求的矩阵， 2Uk，Vk为迭代用的列向量。 3最后的最大特征值maxLamda由最后一次的max(Uk)-----求Uk中的绝对值最大的元素的绝对值.所决定。 而maxLamda所对应的特征向量由最后一次迭代的Vk所决定. 主要的想法就是先选一个不为0的初始向量U0!=0，然后按下面的式子迭代。 -matrix eigenvalue of the largest power France. For engineering calculation, Matrix eigenvalues and eigenvectors are very important and common data, here is the power law is a common solution, using the iterative thinking. Symbol : 1A of the question for the matrix, 2Uk, Vk iteration of the column vector. The final three largest eigenvalue maxLamda from last max (uk Hoffmann for the uk the largest absolute value of the absolute value of the element. by decision. While maxLamda corresponding eigenvectors from the last iteration of Vk decision. The main idea was first choice not one of the initial vector 0 U0! = 0, then by the following formula iteration.

人脸识别，是基于人的脸部特征信息进行身份识别的一种生物识别技术。用摄像机或摄像头采集含有人脸的图像或视频流，并自动在图像中检测和跟踪人脸，进而对检测到的人脸进行脸部的一系列相关技术，通常也叫做人像识别、面部识别。

把1到n的数放在一个环里,使相邻的2个数和是质数.用递归做-n to put a few on a loop, so that the two neighboring and the number is prime. Using recursive do

这是一个列主元消去法的VC程序，是计算方法中的一个重要的算法的编程。-This is a main-element method to eliminate the VC process, the calculation of an important programming algorithm.

逻辑运算类：支持与或非逻辑运算，支持括弧 ( ) 改变运算先后。-Logic operations categories: support and non-logical operations, support brackets () to change the computing successively.

应用背景In generated networks, selecting one node as source and some nodes as receivers in random (single source multicast network), we make performance test for network coding based multicast route algorithm (we put forward it ourselves, corresponding to high multicast rate and low multicast rate), shortest path distribution tree route algorithm (DIJKSTRA algorithm) and Maximum-rate distribution tree route algorithm (PRIM algorithm), respectively. 关键技术 In this program, two things have been done: (1) generate random topologies (2)performance test。   network coding cap:     5.0000  dijk cap:     5.0000  prim cap:     5.0000  network coding res1:     0.0000--res2:    12.0000  dijk res:    14.0000  prim res:    14.0000  average degree:     2.

六自由度工业机器人直线轨迹插补算法源码，对于从工件生成的离散路径点，可通过该程序进行直线拟合以及 相应路径更精确插补，精插补点间隔（插补步长）可自行设置。