ISODATA-MATLAB

用matlab做面积法系统辨识,matlab源码，经过测试，程序辨识准确率高。(use matlab to sovle the problem of system identification by using area method)

相位移频键控的维特比解码 64状态 比较 回溯(QPSK viterbi decode)

PID比例控制，用对水箱的液位的控制，通过调节阀的流量改变，来控制液位稳定(pid)

关于眼图的一个程序，里面有相应的图，以及程序，供大家学习参考(eye pattern)

面板数据模型matlab，可以直接编译运行(Panel data model matlab, can direct the compiler to run)

用matlab编写的RBF神经网络用于分类与回归的程序(Prepared using matlab RBF neural network for classification and regression procedures)

采用AR法模拟风速，计算时间较短，但精度相对于谐波叠加法略差(AR method using simulated wind speed, the calculation time is shorter, but the accuracy is slightly worse relative to the harmonic superposition)

三角分解 输入变量为一个n阶非奇异方阵A,输出的结果为该矩阵的直接分解矩阵(Input variables for a n-order non-singular matrix A, the output of the direct decomposition of the matrix for the matrix)

详细讲解了25个MATLAB图像与视频处理实用案例（含可运行程序），涉及雾霾去噪、答题卡自动阅卷、肺部图像分割、小波数字水印、图像检索、人脸二维码识别、车牌定位及识别、霍夫曼图像压缩、手写数字识别、英文字符文本识别、眼前节组织提取、全景图像拼接、小波图像融合、基于语音识别的音频信号模拟灯控、路面裂缝检测识别、视频运动估计追踪、Simulink图像处理等多项重要技术(Explain in detail the 25 MATLAB image and video processing utility case (including operating procedures), involving haze denoising, sheet automatic marking, lung image segmentation, wavelet digital watermarking, image retrieval, face recognition, license plate location and two-dimensional code recognition, image compression, Hoffman handwritten digit recognition, character recognition, text English anterior segment tissue extraction, panoramic image mosaic, image fusion, light control, analog audio signal of speech recognition pavement crack detection and identification, video tracking, motion estimation of Simulink image processing technology based on a number of important)

In this paper, we show how support vector machine (SVM) can be employed as a powerful tool for $k$-nearest neighbor (kNN) classifier. A novel multi-class dimensionality reduction approach, Discriminant Analysis via Support Vectors (SVDA), is introduced by using the SVM. The kernel mapping idea is used to derive the non-linear version, Kernel Discriminant via Support Vectors (SVKD). In SVDA, only support vectors are involved to obtain the transformation matrix. Thus, the computational complexity can be greatly reduced for kernel based feature extraction. Experiments carried out on several standard databases show a clear improvement on LDA-based recognition