Matlab7

matlab functions matlab functions matlab functions

这是一个不错的资料是关于cdt2000模块的资料可以通过资料查询很多关于这个模块的知识(This is a good information regarding cdt2000 module information through data queries can be a lot of knowledge on this module)

交互式多模型卡尔曼滤波算法,适应于道路交通和航迹跟踪问题(Interactive Multi model kalman filter algorithm )

程序包括EFICA，WASOBI及两者的结合，可以快速实现盲源分离。(The program includes EFICA, WASOBI and their combination, can quickly achieve blind source separation.)

该演示中包含的代码演示如何氡相似的功能，可以用来提高（以及部分）在Connectome电磁图像单元格边界。 请举出下列文件如果您发现此代码有用： Ritwik库马尔，阿梅里奥五雷纳和Hanspeter Pfister说“氡样的特点及其应用Connectomics”，接受，IEEE计算机学会研讨会在生物医学图像分析（MMBIA）2010年数学方法 http://seas.harvard.edu/〜 rkkumar radonLikeFeaturesDemo.m：演示边缘增强用氡相似的功能， getEdgeFeatures.m：助手为高斯函数的二阶导数滤波器 makeBarFilters.m：助手为高斯函数的二阶导数滤波器 sample.png：示例图像在此演示使用(The demo included in the code demonstrates how Radon-Like Features can be used to enhance (and segment) cell boundaries in Connectome EM images. Please cite the following paper if you find this code useful: Ritwik Kumar, Amelio V. Reina & Hanspeter Pfister, “Radon-Like Features and their Application to Connectomics”, accepted, IEEE Computer Society Workshop on Mathematical Methods in Biomedical Image Analysis (MMBIA) 2010 http://seas.harvard.edu/~rkkumar radonLikeFeaturesDemo.m : Demonstrates edge enhancement using Radon-Like Features getEdgeFeatures.m : Helper function for Gaussian Second Derivative Filters makeBarFilters.m : Helper function for Gaussian Second Derivative Filters sample.png : Sample image for use in this demo )

利用K近邻法实现数字识别算法。误差小，识别效率高，网络训练速度快。(K-nearest neighbor algorithm, digital identification algorithm. Error is small, high recognition efficiency and speed of network training.)

基于改进LDA和CNN的网络入侵聚类,该文件为matlab版本的原代码(Based on the improved LDA and CNN network intrusion clustering)

小波分析最基础的、具有导向性的东西，值得(Wavelet analysis of the most basic, with the east-west-oriented, it is worth)

beamforming theory from the renovable engineers

产生零均值单位方差高斯白噪声的1000个样点,估计随机过程的自相关序列(units have zero mean Gaussian white noise variance to the 1,000-point, it is estimated that the random process autocorrelation sequence)