％ 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.
標簽： Z. direction Computes function
上傳時間： 2014-01-24
上傳用戶：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 $
標簽： DEFINEV derivative distances Scaling
上傳時間： 2013-12-24
上傳用戶：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.
標簽： frame Tracking provided changes
上傳時間： 2015-11-17
上傳用戶：zgu489
圖像處理的關于Snakes : Active Contour Models算法和水平集以及GVF的幾篇文章，文章列表為： [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 學習圖像處理的人必看的幾篇文章
標簽： Contour Snakes Active Models
上傳時間： 2014-01-15
上傳用戶：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.
標簽： Carlo Monte algorithm sampling
上傳時間： 2013-12-02
上傳用戶：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.
標簽： frequency function provides response
上傳時間： 2014-01-20
上傳用戶：qunquan
用于汽車巡航控制系統(tǒng)的模糊控制算法，以及如何利用梯度下降法和卡爾曼濾波來優(yōu)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 .
標簽： 汽車巡航 控制系統(tǒng) 模糊控制 算法
上傳時間： 2016-09-07
上傳用戶：xiaodu1124
壓縮包里包含了無約束優(yōu)化問題常用的幾種求解方法的源程序：變量輪換法（variable_rotation.m）、最速下降法（steepest_descent.m）、修正牛頓法（modified_newton.m）、共軛梯度法（conjugate_GRADIENT.m）。另外，coefficient_matrix.m為目標函數(shù)系數(shù)獲得矩陣，minval.m為最小值計算函數(shù)，GRADIENT.m為梯度計算函數(shù)
標簽： variable_rotation steepest_descent modified_newt 源程序
上傳時間： 2017-01-01
上傳用戶：ztj182002
BP 神經(jīng)網(wǎng)絡的基本思想：信號的正向傳播+誤差的反向傳播。 ? 信號的正向傳播：輸入樣本從輸入層傳入，經(jīng)各隱層逐層處理后，傳向輸出層。 ? 誤差的反向傳播：將輸入誤差以某種形式通過隱層向輸入層逐層反傳，并將誤差分攤給各層的所有單元，從而獲得各層單元的誤差信號來作為修正各單元權(quán)值的依據(jù)。 BP算法屬于δ學習規(guī)則類，這類算法被稱為誤差的梯度下降（GRADIENT Descent）算法。 在此分類器中，本文選擇3層BP神經(jīng)網(wǎng)絡算法。隱含層節(jié)點數(shù)為3。
標簽： BP神經(jīng)網(wǎng)絡 非線性 分類器
上傳時間： 2017-05-31
上傳用戶：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
標簽： Convolutional Networks Neural Guide to
上傳時間： 2020-06-10
上傳用戶：shancjb
