为遗传算法提供基类，该基类将评价函数值直接作为适合度，采用新个体直接替换老个体的整体再生法...

The code on this diskette has been organized by chapter. Each subdirectory containing listing has a readme.txt describing program usage and any relevent file formats. The readme will also describe how to compile the target programs under IBM OS/2 (TM) using the using C Set++ compiler (TM). All programs on this diskette have been compiled and tested in this environment. The majority of programs on this diskette should port to other environments with only minor adjustments. The exception to this are those programs which utilize presentation manager for graphical display of program data. In particular, the grid1 programs in chapt4ackprop and chapt3 fall into this latter category.

BP神经网络用于函数拟合与模式识别的Matlab示例程序,说明很详细,并且代码可以直接拿来用-BP neural network for function fitting and pattern recognition procedures Matlab examples to illustrate in great detail, and the code can be directly used by

本人编写的incremental 随机神经元网络算法，该算法最大的特点是可以保证approximation特性，而且速度快效果不错，可以作为学术上的比较和分析。目前只适合benchmark的regression问题。 具体效果可参考 G.-B. Huang, L. Chen and C.-K. Siew, “Universal Approximation Using Incremental Constructive Feedforward Networks with Random Hidden Nodes”, IEEE Transactions on Neural Networks, vol. 17, no. 4, pp. 879-892, 2006. -I prepared by incremental random neural network algorithm, which is characterized by the largest approximation properties can be guaranteed, and fast good results can be used as an academic comparison and analysis. The current benchmark is only suitable for the regression problem. Specific effects may refer G.-B. Huang, L. Chen and C.-K. Siew,

内附两个源代码，其中一个源代码是识别程序的，另一个是一个矩阵类库的，所以一并附上。 ~..~因为已经保存了训练好的网络权值，所以第一次使用识别程序时，可以直接运行Release目录下的可执行文件，然后对图片目录中的测试图片进行读入、识别。 当然读者也可以自行用训练样本训练网络，不过要特别注意训练样本的选择，否则可能造成识别率很低。训练样本选择的原则是，尽可能的有代表性，在训练时间不至于太长的情况下，训练样本数目尽量多。 -containing 2 source code, which is a source identification procedures, and the other is a matrix class library. So accompanied. ~ ~ .. Has been preserved because of the trained network weights, the first use of the identification procedures, can be run directly Release directory executable file, Then the catalog of tests conducted read into the picture, recognition. Certainly readers can use their own training samples training network, but special attention should be paid to the choice of training samples, Otherwise, it could cause a very low recognition rate. Training sample selection principle is that as far as possible be representative, in the training of the time not too long under the circumstances, the number of training samples as much as possible.

基于canny的灰色预测和模式识别,标定的一种做法是：在一块平板上画若干个点， 同时用这两部相机照相，分别得到这些点在它们像平面上的像点，利用这两组像点的几何关系就可以得到这两部相机的相对位置。然而，无论在物平面或像平面上我们都无法直接得到没有几何尺寸的“点”。实际的做法是在物平面上画若干个圆（称为靶标），它们的圆心就是几何的点了。而它们的像一般会变形，如图1所示，所以必须从靶标上的这些圆的像中把圆心的像精确地找到，标定就可实现。-Based on the gray canny prediction and pattern recognition, calibration approach is to: a plate painted in a number of points, at the same time with these two cameras camera, respectively, as these points in their plane as points, the use of these two groups geometric relationship between the image point can be that the relative positions of two cameras. However, in terms of objects such as plane or plane, we can not directly be no geometry point. Actual practice is complex plane painted in a number of round (known as targets), their geometric center of a circle is the point. And they generally like deformation, as shown in Figure 1, it must target these round, like in the center of a circle of like to find precise, calibration can achieve.

一个运用遗传算法来训练神经网络的程序-use of a genetic algorithm to train the neural network process

这是一个feret数据库,对做模式识别的人来说非常有用的库.-液feret 菘, 模 式 识 说 浅玫目.

异步电动机的模糊控制的仿真,大家一起研究.-Fuzzy control induction motor simulation, we study together.

模式识别的书，介绍的论文模式识别的书，介绍的论文-Pattern Recognition book on pattern recognition paper book, the thesis introduction

exmat - The Expression Template Matrix Library,是矩阵运算模板库-exmat- The Matrix Expression Template Library, yes matrix calculation Template Library