资源列表
neural-PID.zip
- 单神经元PID的源程序，此程序简单易懂里面附有详细说明，对于初学者很管用。,Single neuron PID of the source, this procedure is simple to understand which accompanied by detailed notes, for beginners very useful.
GA3threshold
- 针对Bmp图片，采用多阈值最大化图像的信息熵为目标，采用多目标遗传算法（源码全）对图像进行多阈值变换，实现对图像的多阈值分割。（VC6.0实现）-Bmp images using multi-threshold image information entropy maximization as the goal, the use of multi-objective genetic algorithm (source full) image multi-threshold transform,
BP
- 先选择500个数据进行学习，学习方法是采用神经网络的BP算法，然后对未参与学习的数据进行数字识别-Select the first 500 data learning, learning neural network BP algorithm, then digital identification data that did not participate in the learning
RSNA
- 待辨识对象参数a=[1 -1.5 0.7] b=[1 0.5] 输入采用长度L=400的白噪声序列，输出 ，输入和输出数据均含不相关随机噪声，ρ（k）=1/k。利用上述递推公式，辨识系统参数。-To identify the object parameters a = [1-1.5 0.7] B = [1 0.5] Input the length L = 400 white noise sequence, the output, the input and output data
shijuan
- 很好的学习资料，智能信息处理的相关知识，值得下载-Good learning materials, knowledge of intelligent information processing, it is worth to download
GA2
- 基于遗传算法求解多元函数优化问题，程序理论使用变量数目不限制-Based on genetic algorithm to solve the multi-function optimization problem, the use of the procedural theory does not limit the number of variables
RBF
- 径向基神经网络（RBF网络）的三种学习算法实现：随机选取中心法、自组织选取中心和梯度训练算法-Three radial basis function neural network (RBF) learning algorithm: randomly selected center method, self-organizing selection center and gradient training algorithm
M1--12--Study-of-Major-Issues-and-Their-Impact-on
- Study of Major Issues and Their Impact on DVR System Performance
cmean
- 这个算法的基础是误差平方和准则，为了要得到最优结果，首先要对样本集进行划分，一般的做法是先选择一些代表性的点作为聚类的核心，然后把其余的点按照某种方法分到各类中去。-This algorithm is based on error sum of squares criterion, in order to get the optimal results, first of all to sample set division, the general practice is to choose
GA1threshold
- 针对Bmp图片，根据以最大图像的信息熵为目标，采用遗传算法（源码全）对图像进行单阈值变换，实现对图像的阈值分割。（VC6.0实现）-Bmp picture, according to the target maximum image information entropy, genetic algorithm (source all) single threshold transform image threshold segmentation of images. (VC6.0 achieve
GAPBP
- 用遗传算法优化BP神经网络的初始权值，并用于非线性系统辨识。-BP neural network using genetic algorithms to optimize the initial weights, and for the identification of nonlinear systems.
MLS
- 当信噪比较大时，采用广义最小二乘法可能会出现多个局部收敛点，解决这个问题的方法可用多级最小二乘法，一般来说多级最小二乘法包含三级辨识过程。利用输入输出数据，通过多级最小二乘法，可分别求的辅助模型，过程模型和噪声模型的参数估计值。在高噪声的情况下，多级最小二乘法明显优于广义最小二乘法，其收敛点唯一。-When the signal noise ratio (SNR), and by use of the generalized least squares method may appear mor