project22

在用混沌理论和神经网络进行短期负荷预测时，神经网络的输入的选择至关重要，该程序用matlabl实现了基于混沌时间序列的嵌入维数的选择(using chaos theory and neural networks for short-term load forecasts, the neural network is essential to choose an input, The procedure used matlabl achieved a chaotic time series based on the embedding dimension of choice)

BPAdaboost，包括主程序，测试集等等(BPAdaboost, including the main program, test sets, etc.)

Wavelet based compression coding method. Allows the user to compress an image to upto 80 using wavelet based compression method.

简化的ANN程序，程序代码很少，但是预测效果相当(a simplified Ann algorithm with good performance)

说明：  本程序为BP算法在一个简单分类中的应用,主要针对异或问题,有助于对本算法的理解()

2D FDTD realization with Near-to-Far field transformation

LSB匹配算法的实现。目前大部分都是LSB替换算法，较新的图像LSB匹配算法的嵌入和恢复我没有在其它地方找到。因此，自己写了一个，可以作为毕业设计的内容。matlab编写，经过充分测试，可用。(Image LSB Matching Algorithm,implementated using matlab )

追赶法求解矩阵；牛顿插值函数，三次样条插值函数仿真龙格库塔现象；二分法求解根与牛顿迭代，弦割法，简单迭代法的程序(Section three order spline interpolation function Newton interpolation Iterative method)

MATLAB编写的，基于Newmark-beta法的求解微分方程的解法。(method to solve the differential equation,based on the Newmark-beta method.)

该源代码是模式识别课程作业。通过K平均算法实现聚类的目的。效果比较好。另外本人还设计了等高线法，用于聚类，思想较好，效果一般，但是值得参考。(This code is a Pattern Recognition homework. It use Kmeans to clustering, and get a good result. It s good for a reference.)