guangxue-yanshe

本源码是在matlab环境中进行连续小波变换，不用调用matlab中的小波函数。(The source code is in the matlab environment, continuous wavelet transform, do not call matlab in the wavelet function.)

重构Lorenz 时间序列的相空间图，并作分析(Lorenz time series reconstructed phase space, and for analysis)

一种加权支持向量机的算法，主要描述一种加权支持相机的数学描述，参看此文可以更深入理解加权支持向量机(Fast Weighted SVM training Algorithm)

matlab的两本些学习书籍，一本是和C混合变成，一本是图像处理(matlab learning of two more books, one is mixed into and C, an image processing)

The Gaussian mixture probability hypothesis density filter (GM-PHD Filter) was proposed recently for jointly estimating the time-varying number of targets and their states a noisy sequence of sets of measurements. ( The Gaussian mixture probability hypothesis density filter (GM-PHD Filter) was proposed recently for jointly estimating the time-varying number of targets and their states a noisy sequence of sets of measurements .)

matlab访问串口的程序，包括：打开、关闭和清除串口对象；读串口、写串口；串口的wAxzFSp数据流控制；还包括：怎么用matlab发送手机短信的nPDMteb例程；这是学习matlab访问系统硬件的原理。(Matlab serial access program, including: open, close, and clear the serial port object reading the serial port, serial port write serial wAxzFSp data flow control also include: how to use matlab to send SMS wAxzFSpl routines this is to learn matlab access principle of the hardware system.)

几个有关lms的matlab程序，包括一维lms算法，多维lms算法，以及一个使用lms算法实现自适应去噪操作的例子(Several lms matlab procedures, including the one-dimensional LMS algorithm, multi-dimensional LMS algorithm, as well as a example of using the LMS algorithm adaptive de-noising operation)

SVM with K-fold cross validation

the baysian distance metric learning code

Otsu-最大类间方差Matlab代码 最大类间方差法原理：最大类间方差法是由日本学者大津于1979年提出的,是一种自适应的阈值确定的方法,又叫大津法,简称OTSU。它是按图像的灰度特性,将图像分成背景和目标2 部分。背景和目标之间的类间方差 ％ 大,说明构成图像的2 部分的差别越大,当部分目标错分为背景或部分背景错分为目标都会导致2 部 ％ 分差别变小。(Otsu- Otsu Matlab code Otsu Principle : Otsu law by Japanese scholars in the large-1979's, is an adaptive threshold established methodology, also known as big-law, referred to Otsu. It is based on image characteristics of the gray image can be divided into the background and objectives of Part 2. Background and objectives between the between-class variance%, which meant that the two images form part of the greater difference, When some background into the wrong target or part of the background into the wrong target will lead to sub-2% difference becomes smaller.)