人脸特征点检测

多目标优化是指在约束条件下有两个或两个以上的优化目标，而且这些目标之间相互矛盾，不能同时达到最优，也就是说，一个目标的最优往往是以牺牲其它目标作为代价的，因此多目标优化问题存在多个最优解，这些解之间无法比较优劣，统称为Pareto最优解。带精英策略的快速非支配排序遗传算法(Nondominated Sorting Genetic Algorithm II，NSGA-II)是目前应用较为广泛的一种多目标算法。本案例将对MATLAB自带的改进的NSGA-II进行讲解，并举例说明其应用。

应用背景关键技术均值滤波的想法是简单的替换每个像素值与平均图像（`平均”）邻国的价值，包括本身。这消除了其像素值的影响具代表性的环境。像其他部分是围绕一个核心，这代表形状和邻域大小进行采样时计算平均。

关于评价插值多项式 P (N + 1)不同的数字 X(0)，......，在数字 X X(N) 对于函数 F： 输入： 数字 X(0)，......，X(N) 为 XX(1)，......，XX(N+1) ； 数字 X ；值的 F 作为 Q 的第一列，或者可能如果函数 F 提供计算。 输出： 与 P(X) 表 Q = Q(N+1,N+1)。

应用背景This paper considers the problem of interference mitigation in multi-cell multi-antenna time division duplex (TDD) wireless systems for downlink transmission. An efficient way to obtain channel state information (CSI) at the base station is by using uplink pilots and reciprocity of the downlink channel. At the same time, it has been shown that pilots from different cells contaminate each other, resulting in corruption of precoding matrices used by base stations, and high inter-cell interference. This paper studies the effects of shifting the location of pilots in time frames used in neighboring cells, and its effectiveness in obtaining better channel estimates, and, thereby, inter-cell interference reduction.关键技术Wireless cellular network designs with multiple antennas have been adopted by several modern wireless systems, which aim for high throughput as well as high

方程求根_加速迭代法是一个用C++编写的程序-equation Root _ accelerated iterative method is a C preparation procedures

采用opencv进行图片读取和处理，c++实现SSIM算法，评估图片质量，工程在VC6.0开发平台实现，工程中包含opencv所需的头文件和库文件。SSIM算法是最经典的图像质量评价算法，c++源代码可以帮助学习，以供参考。

C++ net time 读取网洛时间-C net Luo time reading time

matlab的数学物理方程数值算法源程序。这是"Numerical Methods for Physics"第二版的matlab源程序。-Matlab mathematical physics equations numerical algorithm source. This is the "Numerical Methods for Physics," the second version of the Matlab source.

Entropy Based Subspace Clustering for Mining Data - ENCLUS - a new version of PROCLUS algorithm for clustering high dimensional data set.-Entropy Based Subspace Clustering for Mining Data- ENCLUS- a new version of PROCLUS algorithm for clustering high dimensional data set.

一维伽辽金型无网格法MATLAB程序 无网格方法采用基于点的近似，可以彻底或部分地消除网格，不需要网格的初始划分和重构，不仅可以保证计算的精度，而且可以大大减小计算的难度。然而，由于目前的无网格近似一般没有解析表达式，且大都基于伽辽金原理，因此计算量很大，要超出传统的有限元法；另外，无网格近似大都是拟合，因此对于位移边界的处理比较困难，多采用拉格朗日乘子法处理。-1维伽Galerkin-type meshless method MATLAB procedures meshless method based on the point of approximation, can be completely or partially eliminate the grid, the grid does not require the initial division and reconstruction, not only can guarantee the accuracy of the calculation, and can greatly reduce the difficulty of calculation. However, due to the current meshless approximation there is no analytic expression in general, and mostly based on the Galerkin principle, the calculation of a large volume, it is necessary to go beyond the traditional finite element method In addition, the meshless approximation are fitted, so displacement of the border more difficult to deal with, the use of Lagrange law.