Speech-recognition

An iterative QuRecursive method (using q-recursive) to generate Zernike

Após o pré-processamento (prep), os dados sã o reduzidos pelo algoritmo SVD e calcula-se os Scores (T) e os Loadings (P). Posteriormente, plota-se T e P, na forma de gráficos biplots

admm，压缩感知的重构程序，内有仿真信号(admm，compressive sensing)

牛顿迭代法，用于对通用多项式的计算，可自己设置处置，精度等(Newton iterative method for the calculation of general polynomials can set up their own disposal, precision)

基于CKF的多模型滤波的英文文献及其代码，能够正确运行。是学习滤波及目标跟踪的好资料。(CKF based on Multiple Model Filtering in English literature and code, to run correctly. Learning and target tracking filter good information.)

能够有效比较FAN和TDOA和泰勒级数展开法三种不同算法在三维定位的优劣性！(Can effectively compare FAN and TDOA and Taylor series expansion method in three different algorithms in three-dimensional positioning and inferiority!)

这是一个说话人识别的程序，其中有十个说话人，识别率可达到90 。执行时，执行“laststep.m”(This is a speech recognition program, including dozens of speaker recognition rate can reach 90 . Implementation, implementation " laststep.m" )

The basic face recognition gives as results the ID of nearest person present in database

hough变换的一个实例，用matlab语言实现。(hough transform.)

包括一些Probability_and_Stochastics 蒙特卡洛算法等程序。。。(Including some Probability_and_Stochastics Monte Carlo algorithms and other procedures. . .)