GA_OPT

说明：  用MATLAB编的 谱估计里的伯格算法 效果很好(Made use of MATLAB in the Burg spectral estimation algorithm works well)

神经网络控制工具箱。包括最优控制，反馈线性化控制，预测控制，前馈控制等。内有说明文件readme.txt(neural network control toolbox. Including optimal control, linear feedback control, predictive control, feedforward control. Have documentation readme.txt)

在1x1的单位矩形中随机部署传感器节点，而且假设每个节点的通信半径一样，要求利用蒙特卡洛算法进行1000次试验分别模拟出连通率随节点数增加以及通信半径增加的变化趋势。(Random units deployed in 1x1 rectangular sensor nodes, and assumed that the communication radius of each node, requires the use of Monte Carlo algorithm to simulate the 1000 trials were increased with the number of nodes increased connectivity and communication radius increasing trends.)

MATLAB工具箱——libsvm-3.17版本，开始学习~(MATLAB Toolbox - libsvm-3.17 version, start learning~)

一 维 时 域 有 限 差 分 法 MATLAB 代 码(One-dimensional finite difference time domain method MATLAB code)

clc close all clear all I=imread( desert.jpg ) [r c]=size(I) bs=16 Block Size (8x8) nob=(r/bs)*(c/bs) Total number of 8x8 Blocks Dividing the image into 8x8 Blocks kk=0 for i=1:(r/bs) for j=1:(c/bs) Block(:,:,kk+j)=I((bs*(i-1)+1:bs*(i-1)+bs),(bs*(j-1)+1:bs*(j-1)+bs)) end kk=kk+(r/bs) end Accessing individual Blocks figure imshow(Block(:,9,1)) This shows u the fist 8x8 Block in a figure window figure imshow(Block(:,:,2))

matching pursuit 算法，matlab实现，信号稀疏分解方式的一种，适合与信号在超完备的原子库中稀疏分解。matchingpursuit.m(A matching pursuit algorithm. It fits data from a set of dictionary elements by orthogonally projecting the data onto elements of the dictionary of vectors such that the energy in each projection is maximized. In constrast to G-S, the residual is orthogonalized, not the basis vectors)

实现ZOOM-FFT的功能，对频谱进行了10倍的细化。(realize ZOOM-FFT fanction)

坐标转换（四参数） [row col]=size(source) for i=1:row u(i)=source(i,2) v(i)=source(i,3) x(i)=source(i,4) y(i)=source(i,5) B(2*i-1,1)=1 B(2*i-1,2)=0 B(2*i-1,3)=u(i) B(2*i-1,4)=-v(i) B(2*i,1)=0 B(2*i,2)=1 B(2*i,3)=v(i) B(2*i,4)=u(i) L(2*i-1,1)=x(i) L(2*i,1)=y(i) end X=inv(B *B)*(B *L) Txy=[X(1) X(2)] xzjz(1,1)=X(3) xzjz(1,2)=-X(4) xzjz(2,1)=X(4) xzjz(2,2)=X(3) [row1 col1]=size(target) for i=1:row1 u1(i)=target(i,2) v1(i)=target(i,3) d=[u1(i) v1(i)] md=Txy+xzjz*d end end (Coordinate conversion： [row col]=size(source) for i=1:row u(i)=source(i,2) v(i)=source(i,3) x(i)=source(i,4) y(i)=source(i,5) B(2*i-1,1)=1 B(2*i-1,2)=0 B(2*i-1,3)=u(i) B(2*i-1,4)=-v(i) B(2*i,1)=0 B(2*i,2)=1 B(2*i,3)=v(i) B(2*i,4)=u(i) L(2*i-1,1)=x(i) L(2*i,1)=y(i) end X=inv(B *B)*(B *L) Txy=[X(1) X(2)] xzjz(1,1)=X(3) xzjz(1,2)=-X(4) xzjz(2,1)=X(4) xzjz(2,2)=X(3) [row1 col1]=size(target) for i=1:row1 u1(i)=target(i,2) v1(i)=target(i,3) d=[u1(i) v1(i)] md=Txy+xzjz*d end end )

Description of the file: template matching source of Chamfer matching algorithm. It s OpenCV implementation