matlab 编写的小波分解程序,进行模式识别-the preparation of the wavelet decomposition matlab procedures, pattern recognition

平时学习时搜集的算法书,聚类,神经网络,蚁群算,遗传算法-some e-books about the algorithm

神经网络C程序的第9章-neural network C procedures for the Chapter 9

我根据书上算法自编的蚁群程序，也是我的中期筛选设计。目录下函数收敛图是用了VC的结果，在matlab里画的-algorithm based on the book written by Ant procedures, and also my mid-screening design. Directory function convergence plan is the result of the VC in Matlab Lane painting

遗传算法解决双变量的函数最优化问题，有按钮的界面，用bc所编，生动模拟遗传进化过程-genetic algorithm to solve the two- variable optimization function, the button interface, using bc prepared by the vivid simulation of the process of genetic evolution

MYCIN不确定推理机的C++骨架程序，运行高效率。-MYCIN uncertainty reasoning machine C++ skeleton program, run efficient.

MFD－多变量系统频域设计工具-MFD- multivariable frequency-domain system design tools

1、编制程序显示印章图像（24位真彩色位图）； 2、读出位图中每一像素点的（R，G，B）样本值； 3、以RGB其中某两个（或三个）为坐标，取一定数量的图像点为分析样本，分析其坐标系中的分布； 4、采用本章学习的方法找到分类判别函数，对这些样本进行分类；（要求首先将印章与底纹区分，如有可能将印章、底纹、签字区分） 5、将分类后的结果标记到原始图像上，检查其效果。 -1, the preparation procedures showed that the seal image (24-bit true color bitmap) 2, read out each bitmap pixel (R, G, B) sample value 3 to RGB which a two (or three ) for the coordinates, take a certain number of image points for the analysis of samples, analysis of its coordinate system of distribution 4, the method used in this chapter to learn to find discriminant function classification of these samples for classification (required first to distinguish between the seal and Shading, if possible, will seal, Shading, signatures to distinguish between) 5, will be the result of classification markings to the original image to check its effectiveness.

标准遗传算法程序，包含详细的问题描述，如果想和大家讨论的话。

微粒群算法与其它进化类算法相类似，也采用“群体”与“进化”的概念，同样也是依据个体（微粒）的适应值大小进行操作。所不同的是，微粒群算法不像其它进化算法那样对于个体使用进化算子，而是将每个个体看作是在n维搜索空间中的一个没有重量和体积的微粒，并在搜索空间中以一定的速度飞行。-Particle swarm optimization algorithm with other similar type of evolution, but also the use of " groups" and " evolution" concept is also based on the individual (particle) size of the operation of fitness. The difference is that unlike other particle swarm optimization evolutionary algorithm as the use of evolution for the individual operator, but as each individual in the n-dimensional search space of a weight and size of the particles, and in the search space the speed of a certain flight.