Jacobian

径向基函数（RBF）神经网络在分类中的应用(Radial Basis Function (RBF) neural network classification)

基于遗传算法的VRP程序，主要是其中几个操作，用c++编写(VRP procedures based on genetic algorithms, several of which operate mainly with the preparation of c++)

利用ar建模 计算ar参数的阶数和ar参数的系数 可以对语音建模(ar modeling)

任意阶zernike多项式的计算，在进行波前拟合时可调用(Zernike any order polynomial terms, during the fitting can be called wavefront)

比较经典的matlab的电力系统潮流计算程序(more classic Matlab power flow calculation procedures)

用matlab实现模糊神经网络，在非线性系统中的应用(Achieved with matlab neural network, nonlinear system)

matlab2010程序教程，里面全是具体算例，比单纯的文字介绍要具体，容易上手。(matlab2010 procedural manuals, which are all specific examples, than simple text description should be specific, easy to use.)

提出了一种求解电磁场有限元 边界元混合法所生成的线性方程组的有效方法 ———内观 法结合多波前法.由于该线性方程组的系数是一个部分稀疏部分满填充的矩阵,为了加速求解,应用内观法将系数矩阵分为 2块,一块是有限元法形成的稀疏矩阵,另一块是边界元法生成的满阵,然后用多波前法求解稀疏矩阵方程,用高斯 约当消去法解满阵方程. 采用该方法,计算了二维多层介质柱体的雷达散射截面.计算结果表明,该方法的计算效率远远高于传统的高斯法.(Proposed for solving mixed finite element boundary element linear equations generated by an effective method of Vipassana--- a combination of multi wave front method. Since the coefficients of linear equations is a part of the sparsely populated part of the full matrix, in order to speed up the solution, application of Vipassana coefficient matrix is ​ ​ divided into two, one is the finite element method for the formation of the sparse matrix, the other is the boundary element method generates the full array, and then use the multifrontal method for solving sparse matrix equation Gauss Jordan elimination method with full matrix equation solution. use this method to calculate the two-dimensional multi-layer dielectric cylinder of the radar cross section. The results show the computational efficiency of this method is much higher than the traditional Gaussian law.)

基于四叉树分解并且使用区域分离和合并的图像分割(based quadtree decomposition and the use of regional separation and integration of image segmentation)

说明：  texture feature extraction using matlab