fsvm

matlab使用垂直Sobel算子，自动选择阈值 [VSFAT Threshold]=edge(f, sobel , vertical ) 边缘探测 figure,imshow(f),title( 原始图像 ), 显示原始图像 figure,imshow(VSFAT),title( 垂直图像边缘检测 ) 显示边缘探测图像 使用水平和垂直Sobel算子，自动选择阈值 SFST=edge(f, sobel ,2) figure,imshow(SFST),title( 水平和垂直图像边缘检测 ) 显示边缘探测图像 使用指定45度角 Sobel算子滤波器，指定阈值 s45=[-2 -1 0 -1 0 1 0 1 2] SFST45=imfilter(f,s45, replicate ) SFST45=SFST45>=2 figure,imshow(SFST45),title( 45度角图像边缘检测 ) 显示边缘探测图像 (matlab I=rgb2gray(I0) J0=double(I) [VSFAT Threshold]=edge(f, sobel , vertical ) figure,imshow(f),title figure,imshow(VSFAT),title)

code to get optical flow using sum of robust differences method

MOSEL IN XPRESS-IVE, IT IS A SOFTWARE FOR OPTIMIZATION

数值分析的EULER法，matlab开发工具箱中的支持向量机，预报误差法参数辨识-松弛的思想。( EULER numerical analysis method, matlab development toolbox support vector machine, Prediction Error Method for Parameter Identification- the idea of relaxation.)

说明：  对轴承的润滑方程进行数值求解仿真，可求解油膜压力，厚度等的分布情况(The lubrication equation of bearing is numerically solved and simulated)

hough变换，matlab编程。Hough变换是一种使用表决原理的参数估计技术。其原理是利用图像空间和Hough参数空间的点－线对偶性，把图像空间中的检测问题转换到参数空间。通过在参数空间里进行简单的累加统计，然后在Hough参数空间寻找累加器峰值的方法检测直线。(hough transform, matlab programming. Hough transform is a vote on the principle of using a parameter estimation techniques. The principle is the point of the image space using the Hough parameter space and line- duality converting the detection problem in the image space to the parameter space. By a simple statistical accumulated in the parameter space, and then in the Hough parameter space to find a method of detecting the peak accumulator line.)

卡尔曼滤波实例，最基础的卡尔曼滤波，适合初学者(Kalman filter instance, the most basic Kalman filter for beginners)

G-Rilling的emd工具箱,很好的完成hht中的黄变换。(G-Rilling s emd toolbox, well done hht of Huang Transform.)

说明：  BESS通常用于负载均衡，调峰，负载转移应用等。 此BESS块从工作空间获取每小时负载曲线（kW）输入，并计算电网和电池使用量输出到工作空间。 必须以两列格式准备负载配置文件，其中第一列是从一天的0小时开始的时间。第二列是以kW为单位的功耗。（例如，将SampleLoadProfile.mat加载到工作空间中）。 模拟运行时间是从一天的0小时开始的每小时单位。例如，为了模拟24小时负载配置文件，模拟运行时间设置为23，一周运行时间设置为167，一个月30天运行时间设置为719和31天运行时间设置为743。 在“求解器”选项下的模型配置参数中，设置为固定步长类型并将固定步长（基本采样时间）设置为1并运行模拟。(BESS are commonly used for load leveling, peak shaving, load shifting applications and etc. This BESS Block takes hourly Load Profile (kW) input from workspace and compute the Grid and Battery usage output to workspace. The load profile has to be prepared in two column format, where the first column is time starting from 0 hour of the day. The second column is power consumption in kW unit. (For example, load the SampleLoadProfile.mat into workspace). The simulation run time is in hourly unit starting from 0 hour of the day. For example to simulate a 24 hours load profile, the simulation run time is set to 23, one week run time is set to 167, one month 30 days run time is set to 719 and 31 days run time set to 743. In model configuration parameters under Solver options, set to fixed-step type and set the fixed-step size (fundamental sample time) to 1 and run the simulation.)

变步长自然梯度算法实现盲源分离，后面还有动量项自然梯度算法(Variable step-size natural gradient algorithm for blind source separation)