fit

立体匹配算法中比较经典的块匹配方法。其中包含了比较有名的ＳＡＤＳＳＤ算法－－－即左右视图中逐一比对像素数值，从而求得最小ｄｉｓｐａｒｉｔｙ　继而得到ｄｅｐｔｈ。(ＳＳＤＳＡＤ　methods present in this file,which are improtant in the stereo matching-----we obtain the disparity with the difference of the intensity between the left and right images )

实现微电网逆变器的平滑切换PQ和VF之间切换，供参考学习交流用(o achieve smooth switching between microgrid inverters PQ and VF switching, for reference, learning and communication.)

时序网络的时间序列分析建模方法，以及Kalman滤波方法的参数估计(complex network time series )

扩展卡尔曼滤波算法用matlab实现。分别计算出观测值和卡尔曼滤波值进行对比。(Extended Kalman filter algorithm using matlab implementation. Observed values were calculated values were compared and Kalman filtering.)

直流电机的直接起动仿真，主要在simulink中(Start to start the simulation of DC motor)

RHis mfiles is for optimlal conr=trol ddassdddd

说明：  若锗中含有一定数量的杂质元素Sb，试根据要求分析杂质浓度与电离度以及电离温度之间的关系： （1）当Sb浓度分别为 和 时，计算杂质99％，90％和50％电离时的温度各为多少？ （2）根据一定杂质类型和杂质浓度，画出电离度和温度的关系图线，并确定半导体处于强电离区（电离度>90％）的温度范围。 设计要求：（1）具有友好输入输出界面； （2）调整输入数据，得出相应结果，并进行分析。 (If the germanium containing a certain number of impurity elements Sb, test analysis in accordance with the requirements of impurity concentration and ionization degree, as well as the relationship between the ionization temperature: (1) When the Sb concentration and time were calculated 99 percent of impurities, 90 and 50 ionization When the temperature of each number? (2) according to certain types of impurities and impurity concentration, ionization degree and temperature to draw the relationship between the graph lines, and to determine the semiconductor in a strong ionization zone (ionization degree)

高斯迭代法求线性方程组，用于数值分析方面，(Gauss-Seidel)

Recursive least squares递归最小二乘算法(Recursive least squares)

说明：  几个神经网络的算法实例，及相关优化，希望对了解神经网络有帮助(Several examples of neural network algorithms, and related optimization, we want to understand the neural network has helped)