I-component-analysys

RELAX谱估计法的matlab源代码，已经通过验证(RELAX spectral estimation method matlab source code has been validated)

神经网络的ELM算法，比传统的BP和SVM都快，而且效果也很精确。运行平台是matlab，本人已经对原始ELM做了修改，可以适应多种函数，而且在数据处理方面自动产生分类矩阵。(ELM neural network algorithm, than the traditional BP and SVM faster, but the effect is also very accurate. Platform is running matlab, I have to do the original ELM been modified to adapt to a variety of functions, but also in data processing is automatically generated classification matrix.)

触发器，eda.在实际的数字系统中往往包含大量的存储单元，而且经常要求他们在同一时刻同步动作，为达到这个目的，在每个存储单元电路上引入一个时钟脉冲（CLK）作为控制信号，只有当CLK到来时电路才被“触发”而动作，并根据输入信号改变输出状态。把这种在时钟信号触发时才能动作的存储单元电路称为触发器(Triggers, eda. The actual digital systems often contain a large number of memory cells, and at the same time they often require synchronized action, to achieve this purpose, the introduction of a clock pulse (CLK) in each memory cell circuit as a control signal , the arrival of the circuit only when CLK was only trigger the operation and change the output state of the input signal. The memory cell circuit that can trigger actions when the clock signal is called a flip-flop)

S4VM 对传统的S3VM 进行了改进。传统的S3VM 基于低密度假设，它试图找到一个低密度的分界线， 也就是更倾向于决策边界穿过特征空间的低密度区域。S4VM 和S3VM 的不同点在于，S3VM 试图把注意力 集中在一个最优的低密度分界线上，而S4VM 则同时关注多个可能的低密度分界线。之所以这样做，是因为 给定一些有标记的点和大量未标记的点，可能存在着不止一个“间隔”较大的低密度分界线（如图2 ），基于 有限的标记样本，很难决定哪个是最好的。虽然这些低密度分界线都与有限的标记样本吻合，但它们之间差 异很大，因此如果选择错了，会有一个很大的损失，导致性能下降。所以S4VM 试图考虑所有可能的低密度 分界线。 (Over the past decade, many improved versions are proposed to solve every different specific problem of semi-supervised support vector machines, for example: to the inefficiency issue, meanS3VM is introduced to performance degenerate problem, S4VM is proposed)

这里有高斯同态滤波，住址滤波，空域高通滤波，下早后均可直接运行(Here are Gaussian homomorphic filtering, address filtering, spatial high-pass filtering, can be run directly under the early post-)

才用CDD对灰度或者彩色图像进行修复的一种方法(Only with the CDD on the grayscale or color image restoration method)

说明：  目前各类粒子群算法的应用、改进、参数设置方法(At present the application of various types of particle swarm optimization, improvement, parameter setting method)

control induction machine by VSI simulink file

it s a sample for initiaiton to matlab

ICP模型配准有二维三维两种方式，简单好用(ICP RIGID EASY 2 DIMENSION 3 DIMENSION)