2011年03月_juliosun

原创 matlab diff（X)

Y = diff(X) calculates differences between adjacent elements of X. Y = diff(X,n) applies diff recursively n times, resulting in the nth difference. Thus, diff(X,2) is the same as diff(diff(X)).

2011-03-23 14:32:00 2079

功能：用于从对象中移除小对象。用法：BW2 = bwareaopen(BW,P)BW2 = bwareaopen(BW,P,CONN)BW2 = bwareaopen(BW,P,CONN)从二值图像中移除所有小于P的连通对象。CONN对应邻域方法，默认为8。例子：originalBW = imread('text.png');imview(originalBW)bwAreaOpenBW = bwareaopen(originalBW,50);imview(bwAreaOpenBW)

2011-03-22 20:31:00 26856 4

原创 matlab中的size函数

matlab中的size函数size：获取数组的行数和列数（1）s=size(A),当只有一个输出参数时，返回一个行向量，该行向量的第一个元素时数组的行数，第二个元素是数组的列数。（2）[r,c]=size(A),当有两个输出参数时，size函数将数组的行数返回到第一个输出变量，将数组的列数返回到第二个输出变量。（3）如果在size函数的输入参数中再添加一项，并用1或2为该项赋值，则size将返回

2011-03-22 14:54:00 4759 1

原创 GUI加菜单 matlab

在OpeningFcn中添加语句: set(gcf, 'menu','figure') 即可在gui中使用标准菜单

2011-03-22 12:59:00 1161

原创 Feedback Nerual Network(六）：Boltzmann Machine

It is a Random NN, is also a kind of Feedback NN It is similar with DHNN in many aspects, big difference is that Boltzmann NN is a Random NN When the temperature parameter T→ 0 ⇒S-Type Function approaches to Binary Function ⇒R

2011-03-15 15:03:00 1054

原创 Feedback Nerual Network(五）：Design CHNN

Difference between CHNN & DHNN: First-order differential equation corresponds to first-order inertial link惯性环节. For DHNN, this formula is yi=si. Usually it is not Binary Function any more here, but use S-type Function or Hype

2011-03-15 14:43:00 770

原创 Feedback Nerual Network(四）：Design DHNN

Asynchronous Mode: # Conclusion: from above calculation we can see Using different adjusting sequence, x(5) can either weakly converge to x(2)

2011-03-15 13:26:00 813

原创 Feedback Nerual Network(三）：Design symetric link weight matrix

Design a DHNN (1) 1) Calculate Link-Weight Matrix 2) Obtain Attractor Calculate Link-Weight Matrix: not non negative defined symmetric.

2011-03-15 12:02:00 566

原创 Feedback Nerual Network(二）：Stability and Attractor

Definition Attractor: If the state x of NN satisfies then x is called as the Stable Point or Attractor of the NN Definition of Energy Function: Energy Function is a function of neuron’s Status & Link-Weight,which as last should be

2011-03-15 11:54:00 604

原创 Feedback Nerual Network(一）：Concept

It requires a certain period of working time, then can reach Stable Status, so it is a Dynamic Network、 Two types of Feedback NN according to the value of output is discrete or continuous DHNN: Discrete Hopfield NN<br /

2011-03-15 11:19:00 655

转载 matlab画任意圆柱

function [Cylinder EndPlate1 EndPlate2] = cylinder3(X1,X2,r,n,cyl_color,closed,lines)%% This function constructs a cylinder connecting two center points % % Usage :% [Cylinder EndPlate1 EndP

2011-03-09 17:06:00 12797 5

原创点光源小孔成像matlab

模拟的r射线，不用考虑散射了。function []=spheresource()h=input('source to collimator:');H=input('source to detector:');r=input('hole diameter:');d1=input('pinhole1 parameter(x y axis):');d2=input('pinhole2 parameter:');d3=input('pinhole3 parameter:');d4=input('pinhole4

2011-03-05 18:18:00 3546

原创神经网络（六）：前馈网络Delta study principle and Hebbian study principle

Delta study principle: Hebbian study principle: Approach study principle:

2011-03-05 18:08:00 1901

原创神经网络的基本知识(五）：前向网络：BP网络

BP网络其实还是前向反馈网络中的一种，她对于处理非线性分类问题有独特的优势。 BP NN can approach to any Non-Linear Mapping Relationship, only if there are enough hide layer & hide nodes. Structure of BP Network is same as multi-Layer Perceptron Network only their Transfer

2011-03-05 17:57:00 2297

原创神经网络基本介绍（四）：前馈网络（下）多层感知机

（四）multi-layer perceptron network Can solve Exclusive Linear-non-classifiable input patterns，这解决了单层感知器的缺点.本质上是由于加了隐藏层的原因。 Each layer equivalent to a Single-Layer Perceptron Networks qth layer forms a nq-1 dimension

2011-03-05 17:34:00 3139

原创神经网络基本介绍（三）：前馈网络（上）M-P model 和单层感知机

上图是前馈网络基本结构，分别介绍几种前馈网络： M-P model，single-layer perceptron network,multi-layer perceptron network, BP(back propagation)(一）M-P modelProposed by McCulloch & PittsConsists of fixed Structure & fixed Link-WeightsRestrain Convex-Touch Weight ⇒ output is 0Excitin

2011-03-05 17:11:00 2265

原创神经网络基本知识（二）：简单分类

前面简单介绍了神经网络的概念，下面进一步了解。一。首先要清楚几个问题，什么时候可以使用？当系统要求不是严格100%输出正确率时。分类（classification)还是回归(regression)？通常状况下是分类，简单的线性分类或是非线性分类。回归是指使估计的曲线尽量接近所有分类点，实际上是分类的特殊情况。确定的(deterministic)还是随机的（stochastic)？weight权重可能有固定的最佳值，或是接近值，这是是确定的。随机的即是不确定的权重值。supervised还是unsupervi

2011-03-05 13:56:00 2564

#+!