D

这是可控金字塔（steerable pyramids）的代码，使用Matlab 编写(This is controllable pyramid (steerable pyramids) code, the use of Matlab to prepare)

cohere(x,y,M) computes the coherence function using successive DFTs of length with a Hanning window and 50 overlap. (The window and overlap can be controlled via additional optional arguments.) We see a coherence peak at frequency cycles/sample, as expected, but there are also two rather large coherence samples on either side of the main peak. These are expected as well, since the true cross-spectrum for this case is a critically sampled Hanning window transform. (A window transform is critically sampled whenever the window length equals the DFT length.)

机器人障碍环境下路径规划问题，采用改进的回退蚁群算法，matlab版源代码，有可视化界面。(Robot obstacle environment path planning problem, using an improved back colony algorithm, matlab version of the source code, visual interface.)

Check this file for sensor

克里金近似模型是一种很好的近似模型，使用非常广泛。(Kriging approximation model is a good approximation model is widely used.)

MCMC方法主要是为了解决有些baysian推断中参数期望E(f(v)|D)不能直接计算得到的问题的。 其中v是要估计的参数，D是数据观察值(The concept consists of two parts:markov chain and monte carlo integration。)

实现矩阵的LU最大主元分解，方便快捷有效，可以查看分解过程，适合初学者(Achieve maximum principal component matrix LU decomposition, convenient and effective, you can view the decomposition process, suitable for beginners)

human pose estimation

Simulation of channel estimation and equalization for WiMAX PHY layer in simulink (pdf)

共轭梯度法（Conjugate Gradient）是介于最速下降法与牛顿法之间的一个方法，它仅需利用一阶导数信息，但克服了最速下降法收敛慢的缺点，又避免了牛顿法需要存储和计算Hesse矩阵并求逆的缺点，共轭梯度法不仅是解决大型线性方程组最有用的方法之一，也是解大型非线性最优化最有效的算法之一。 在各种优化算法中，共轭梯度法是非常重要的一种。其优点是所需存储量小，具有步收敛性，稳定性高，而且不需要任何外来参数(Conjugate Gradient method (Conjugate Gradient) is between the steepest descent method between Newton method and a method, it only USES a derivative information, but overcome the steepest descent method slow convergence of weakness, but also avoid the Newton law needs to storage and computing Hesse inverse matrix and shortcomings, Conjugate Gradient method is not only solve linear equations with most of the large method, and also one of the most effective solution large nonlinear optimization of one of the algorithm. In all kinds of optimization algorithm, the conjugate gradient method is very important. Its advantage is the storage capacity needed, it has small step convergence, high stability, and doesn t require any exotic parameters numerical experiment, this is the modern scientific computing of the answer above problem sets)