nosGUI

pid仿真控制器，设计了一个最简单的pid仿真器，并将其模块化(pid emulator)

图像插值，最邻近，双线性，双三次算法函数(Image interpolation, nearest neighbor, bilinear, bicubic algorithm function)

基于信息最大化的盲源分离，采用2路混合超高斯信号，分离效果明显(Based on information maximization blind source separation using 2-way mixed super-Gaussian signal, separation effect is obvious)

matlab非常有用的工具箱，可对非平稳信号进行分析与处理。(matlab toolbox very useful, may be non-stationary signal analysis and processing.)

The program is about buck converter PWM in simulink.

online chat application

用matlab语言描述了列车的牵引、制动等过程，给列车方向研究的伙伴们提供个参考(introduce the method of train traction and brake in the process of its running in the tunnel)

image regognition using matlap and improving. done by me in lab session

说明：  高斯过程在分类中的应用。本程序采用拉普拉斯逼近基于高斯过程的贝叶斯分类。(Gaussian process in the Classification. This procedure uses the Laplace approximation of Gaussian processes based on Bayesian classification.)

利用pcnn，每当有一批像素对应的神经元点火，对像素值进行一次修正。第n次点火的所有神经元用矩阵B（n）表示， 已经点火的像素位置标记为‘1’，未点火的标记为‘0’。通过一个3*3的模板滑过B（n），判断若模板内的值全为‘1’或全为‘0’， 则这些像素值不进行处理，否则若模板中心的值为‘1’，则增加该位置的像素值的大小，中心值为‘0’，则减小像素值。该功能由xiugai(B,K)函数实现 Beta取负值来抑制周围的神经元点火，因为输入pcnn（X）的是模糊图像，抑制之后使处理的像素值更加接近原图像的像素值（抑制参数通过试验确定，并不精确） 总之运行结果只是对像素值变化较大的边缘区域进行了像素值的修正。 通过不断调整参数使峰值信噪比有所升高，但是升高得不是太多(Use pcnn, whenever a group of pixels corresponding to the firing of neurons, a correction value for the pixel. Ignition of all the n-th neuron matrix B (n) represents, Already ignition pixel position labeled 1 , unfired labeled 0 . By a 3* 3 templates over the B (n), determines if the value within the template are all 1 or all 0 , These pixel values are not processed, otherwise the value if the template center 1 , then the increase in the size of the pixel values of the position, the center value of 0 , the pixel value is reduced. This function is implemented by xiugai (B, K) function Beta is a negative value around the element to inhibit neuronal firing, because the input pcnn (X) is a blurred image, the pixel values of inhibition after treatment of the pixel value closer to an original image (suppression parameter determined by experiment, is not exact) Results only in short run large changes in the edge region of the pixel values of the corrected pixel values. )