％ COMPDIR Computes a search direction in a subspace defined by Z. ％ Helper function for NLCONST. ％ Returns Newton direction if possible. ％ Returns random direction if gradient is small. ％ Otherwise, returns steepest descent direction. ％ If the steepest descent direction is small it computes a negative ％ curvature direction based on the most negative eigenvalue. ％ For singular matrices, returns steepest descent even if small.
标签： Z. direction Computes function
上传时间： 2014-01-24
上传用户：Thuan
％DEFINEV Scaling vector and derivative ％ ％ [v,dv]= DEFINEV(g,x,l,u) returns v, distances to the ％ bounds corresponding to the sign of the gradient g, where ％ l is the vector of lower bounds, u is the vector of upper ％ bounds. Vector dv is 0-1 sign vector (See ?? for more detail.) ％ ％ Copyright (c) 1990-98 by The MathWorks, Inc. ％ $Revision: 1.2 $ $Date: 1998/03/21 16:29:10 $
标签： DEFINEV derivative distances Scaling
上传时间： 2013-12-24
上传用户：sz_hjbf
Tracking a moving object through several frames, provided changes from frame to frame are on the order of +-(10 + "X Range") pixels in the X direction and +-(10 + "Y Range") in the Y direction is done automatically because of a relatively large area of exploration during the search for an optimal (new) position for a particular control point and a very strong force exerted by large values of the image gradient.
标签： frame Tracking provided changes
上传时间： 2015-11-17
上传用户：zgu489
图像处理的关于Snakes : Active Contour Models算法和水平集以及GVF的几篇文章，文章列表为： [1]Snakes Active Contour Models.pdf [2]Multiscale Active Contours.pdf [3]Snakes, shapes, and gradient vector flow.pdf [4]Motion of level sets by mean curvature I.pdf [5]Spectral Stability of Local Deformations Spectral Stability of Local Deformations.pdf [6]An active contour model for object tracking using the previous contour.pdf [7]Volumetric Segmentation of Brain Images Using Parallel Genetic AlgorithmsI.pdf [8]Segmentation in echocardiographic sequences using shape-based snake model.pdf [9]Active Contours Without Edges.pdf 学习图像处理的人必看的几篇文章
标签： Contour Snakes Active Models
上传时间： 2014-01-15
上传用户：wqxstar
Hybrid Monte Carlo sampling.SAMPLES = HMC(F, X, OPTIONS, GRADF) uses a hybrid Monte Carlo algorithm to sample from the distribution P ~ EXP(-F), where F is the first argument to HMC. The Markov chain starts at the point X, and the function GRADF is the gradient of the `energy function F.
标签： Carlo Monte algorithm sampling
上传时间： 2013-12-02
上传用户：jkhjkh1982
sfrmat is a Matlab function that provides a spatial frequency response* (SFR) from a digital image file containing a slanted-edge feature. The specific edge-gradient algorithm follows the intent of the standard ISO 12233, developed by Technical Committee ISI/TC 42, for resolution measurements for electronic still pictorial cameras.
标签： frequency function provides response
上传时间： 2014-01-20
上传用户：qunquan
用于汽车巡航控制系统的模糊控制算法，以及如何利用梯度下降法和卡尔曼滤波来优化模糊控制器的算法。The files illustrate a simple fuzzy control algorithm as applied to an automobile cruise control system. The files also illustrate how gradient descent and Kalman filtering can be used to optimize the fuzzy controller .
上传时间： 2016-09-07
上传用户：xiaodu1124
压缩包里包含了无约束优化问题常用的几种求解方法的源程序：变量轮换法（variable_rotation.m）、最速下降法（steepest_descent.m）、修正牛顿法（modified_newton.m）、共轭梯度法（conjugate_gradient.m）。另外，coefficient_matrix.m为目标函数系数获得矩阵，minval.m为最小值计算函数，gradient.m为梯度计算函数
标签： variable_rotation steepest_descent modified_newt 源程序
上传时间： 2017-01-01
上传用户：ztj182002
BP 神经网络的基本思想：信号的正向传播+误差的反向传播。 ¡ 信号的正向传播：输入样本从输入层传入，经各隐层逐层处理后，传向输出层。 ¡ 误差的反向传播：将输入误差以某种形式通过隐层向输入层逐层反传，并将误差分摊给各层的所有单元，从而获得各层单元的误差信号来作为修正各单元权值的依据。 BP算法属于δ学习规则类，这类算法被称为误差的梯度下降（gradient Descent）算法。 在此分类器中，本文选择3层BP神经网络算法。隐含层节点数为3。
上传时间： 2017-05-31
上传用户：jplalala
General paradigm in solving a computer vision problem is to represent a raw image using a more informative vector called feature vector and train a classifier on top of feature vectors collected from training set. From classification perspective, there are several off-the-shelf methods such as gradient boosting, random forest and support vector machines that are able to accurately model nonlinear decision boundaries. Hence, solving a computer vision problem mainly depends on the feature extraction algorithm
标签： Convolutional Networks Neural Guide to
上传时间： 2020-06-10
上传用户：shancjb
