GS2

python机器学习预测分析核心算法的例程(Python machine learning prediction and analysis of core algorithm routines)

利用PHOTOSHOP软件可以量测出左右像片各控制点的像素坐标，通过公式可以计算出其像点坐标。再根据给定的控制点的空间坐标，利用直线线性变换解法计算出11个DLT参数即l的系数值以及内方位元素然后解算出外方位元素。另外一种方式是用单像空间后方交会的公式计算出外方位元素。利用单像空间后方交会的来结算影像的外方位元素的基本思想是：以单幅影像为基础，从该影像所覆盖地面范围内若干控制点的已知地面坐标和相应点的像坐标量测值出发，根据共线条件方程，解求该影像在航空摄影时刻的外方位元素。首先建立单文档的程序，建立一个对话框类，在这个类中进行像素坐标、控制点坐标文件的打开与读取，然后利用DLT公式进行外方位元素的解算。 数据检核部分用前方交会算法来进行，首先根据ID找到左右片的像点的同名点，然后计算变换矩阵及系数，根据给定的内方位元素，利用公式解算出像点的控制点坐标，然后和给定的控制点坐标相比较即可进行精度评定。 (PHOTOSHOP software can be measured using a photo of the control points around the pixel coordinates can be calculated by the formula sit out the image point Standard. According to the given coordinates of control points, the use of straight-line method to calculate the linear transformation of 11 DLT parameters that the coefficient l Value and calculate the orientation elements of exterior orientation elements and solutions. Another way is to use a single formula as the calculation of space resection Orientation elements out. Use as space resection alone to settle the image exterior orientation is the basic idea: to single video As the basis of images from the ground covered by a number of control points within the known ground coordinates and image coordinates of corresponding points in the measured values , The realities of collinearity equation, find the image in the time of aerial photography exterior orientation elements. First, the process to establish a single document)

用MATLAB实现模拟立方体的重构,是计算机视觉中三维重构的重要的应用实验(Simulating cube reconstruction with MATLAB is an important application experiment of 3D reconstruction in computer vision.)

based classification (SRC) has been widely used for face recognition (FR). SRC first codes a testing sample as a sparse linear combination of all the training samples, and then classifies the testing sample by evaluating which class leads to the minimum representation error. While the importance of sparsity is much emphasized in SRC and many related works, the use of collaborative representation (CR) in SRC is ignored by most literature.

最新的关于贝叶斯聚焦算法IEEE文章。该算法通过利用信号DOA概率密度函数实现加权的阵列波前差值法，能实现宽带相干信号的DOA估计，并能减小阵列聚焦误差，提高宽带阵列频域波束形成鲁棒性。(In this paper, we present and study a Bayesian focusing transformation (BFT) for coherent wideband array processing, which takes into account the uncertainty of the direction of arrivals (DOAs). The Bayesian focusing method minimizes the mean-square error of the transformation over the probability densityfunctions(pdfs)oftheDOAs,thusachievingimprovedfocusing accuracy over the entire bandwidth)

说明：  算法知识讲解，是一本不错的算法入门级教材的pdf版本(This is a good introduction to algorithm)

Echo cancellation using NLMS adaptive filter

利用风速u,v计算海表风应力以及风应力旋度。(Calculation of sea surface wind stress and wind stress curl.)

说明：  分数傅里叶变换程序，这个是二维分数阶傅里叶变化程序。(Fractional Fourier transform procedure, this is a two-dimensional fractional Fourier change process.)

三角形网格划分, 自由网格和结构网格划分，包括主要的matlab文件和函数，使用main.m 执行(mesh triangle grid using matlab, free mesh and structured mesh. The file include mfiles and functions. The program could be run through main.m)