模糊C均值算法的m文件带自己的解释，希望能给大家一点学习上的帮助。...

Fixed-point for functions of several variables (Function fixed_point_systems(x0, N) approximates the solution of a system of nonlinear equations F(x) = (f1(x), f2(x), ..., fn(x)) = 0 rewritten in the fixed-point form x = G(x) = (g1(x), g2(x), ..., gn(x)). This is done by the user at the bottom of the file. The sub-function G(x) has to be updated accordingly with the given system. The iteration scheme x(k+1) = G(x(k)) is implemented until condition |x(k)- x(k-1)| < tol is met, starting with an initial approxiamtion vector x0. If, after N iterations, the stopping criterion is not reached, a message concerning the iterated x s is displayed. the output consists of iterates of each variables x1, x2, ..., xn in each column. )

说明：  BP网络是通过输入层到输出层的计算来完成的。多于一层的隐含层虽然能在速度上提高网络的训练,但是在实际应用中需要较多的训练时间,而训练速度可以用增加隐含层节点个数来实现,因此在应用BP神经网络进行预测时,选取只有一个隐含层的三层BP神经网络就足够了。(The BP network is completed through the calculation of the input layer to the output layer. Although the hidden layer of more than one layer can improve the training of network in speed, more training time is needed in practical application, and the training speed can be realized by increasing the number of hidden layer nodes. Therefore, when the BP neural network is used to predict, the three layer BP neural network with only one hidden layer is sufficient.)

MultiScale Retinex 色彩恒常性的彩色图像增强算法(Color constancy MultiScale Retinex for Color Image Enhancement Algorithm)

这是HARRIS1988年角点检测的原文。。。非常有参考价值(This is the original HARRIS1988 corner detection. . . Reference value)

matlab 分别利用邻域平均和K点平均对椒盐噪声图像，高斯噪声图像，乘性噪声图像进行处理，比较两者的差别(matlab use neighborhood were average and K-point average for salt and pepper noise image, Gaussian noise images, image processing multiplicative noise, compare the difference between the two)

说明：  该代码包括单隐含层BP和双隐含层BP。建立基于BP神经网络的预测模型，对数据进行随机排列，选取训练样本和测试样本，训练样本训练网络，测试样本进行验证(The code includes single hidden layer BP and double hidden layer BP. Establish a prediction model based on BP neural network, arrange the data randomly, select training samples and test samples, train the training network of training samples, and verify the test samples.)

包括AHP，因子分析，回归分析，聚类分析，关于神经网络控制，利用贝叶斯原理估计混合logit模型的参数，对HARQ系统的吞吐量分析，保证准确无误，是学习通信的好帮手，能量熵的计算。

说明：  设计产生一个连续信号，包含低频，中频，高频分量，对其进行采样，进行频谱分析，使用矩形窗设计不同特性的数字滤波器对信号进行滤波处理，分析所设计滤波器(画出了频率特性曲线)，并对信号进行滤波，画出滤波后信号的时域波形和频谱，并对滤波前后的信号进行对比，同时设计出了一个友好的人机交互界面。(Designed to produce a continuous signal, including low frequency, medium frequency, high frequency components, its sampling, spectral analysis, using rectangular windows of different characteristics of digital filter design the signal filtering, analysis of the designed filter (to draw the frequency characteristic curve), and signal filtering, to draw the filtered signal in time domain waveforms and frequency spectrum, and filtering the signal before and after comparison, the mean time, a friendly interactive interface.)

分数阶傅立叶变换程序 Fractional Fourier Transform(Fractional Fourier Transform)

利用模糊C聚类（FCM）的方法对彩色图像进行分割，期中可以用不通的特征向量来表征图像的特征(Using Fuzzy C-clustering (FCM) method for color image segmentation, mid-term feature vectors can be used unreasonable to characterize the image Feature)