Upload Code
loading-left
loading loading loading
loading-right

Loading

Profile
No self-introduction
codes (314)
Application of genetic algorithm to optimize BP ne
no vote
Application background In the late 80's of the 2O century, multi robot cooperation has become a new type of robot application. On the one hand, due to the complexity of the task, when the single robot is difficult to complete the task, people hope that through the coordination and cooperation between the multi robot to complete. On the other hand, people also hope to improve the efficiency of the robot system in the process of operation by the coordination and cooperation of the multi robot human. In 1943, Maeullocu and Pitts integration of mathematical biophysics and presented the first neuron model. From this, the artificial neural network has experienced the development, stagnation, and then development of the process, now to the mature, in a wide range of applications, which will be artificial neural network technology applied to multi robot cooperation has become a new research field. In this paper, we study the task of J -3, which is controlled by artificial n
njwhat
2016-08-23
0
1
Fault tree simulation analysis based on Monte Carl
no vote
Application background Quantitative calculation of the top event probability is an important aspect of the fault tree analysis, usually with the formula calculation, not only a large amount of calculation, and the calculation process is complicated and cumbersome and error prone. By Monte Carlo simulation method to carry on the simulation of the fault tree model to calculate top event occurrence probability and some other reliability index. Through the example proved. This method is simple and convenient, the result of high precision and reliability of complex system analysis has a very important significance. Key Technology According to the function and working characteristics of digital relay protection system, a quantitative analysis method of dynamic reliability of protection system is proposed. First of all, the establishment of the protection system of the cumulative failure probability, availability and basic components of the probability
njwhat
2016-08-23
0
1
Radial basis function neural network Matlab source
no vote
Application background As is known to all, the BP network is used for function approximation, and the adjustment of weight is negative gradient descent method. The method has limitations, namely, slow convergence and local minimum. Radial basis function network (RBF) is superior to BP network in approximation ability, classification ability and learning speed.Matlab provides a function of four radial basis functions, which are created by the two layers of the neural network, the first layer is a radial base, the second layer is linear layer or layer. The main difference is their weight, even if the function of different threshold or whether there is a threshold.  Key Technology RBF network can approximate any nonlinear function, it can deal with the difficulty in the system, it has good generalization ability, and has a fast learning convergence speed. It has been successfully applied to nonlinear function approximation, time series analysis
njwhat
2016-08-23
1
1
Sakhi Omid "using SVM for face recognition"
no vote
Application background Principal component analysis (PCA) is the main method of feature extraction in face recognition. Support vector machine (SVM) has many advantages, such as small sample, nonlinear and high dimension. The use of kernel function and generalization ability, and the combination of the two. First, using the fast PCA algorithm to extract and select the face image feature vector (SVM). Finally, we use the trained support vector machine (SVM) for face recognition. Key Technology This paper introduces the design of face recognition method classifier and carries on face recognition. SVM classifier is designed, which is used for face recognition, and the experimental results are analyzed. The main steps of the experiment: first of all, the image preprocessing, converted into vector, and then through the ORL algorithm to extract the PCA face database images, the use of SVM toolbox to train the data, and then use the SVM classification m
njwhat
2016-08-23
0
1
Face recognition system using neural network
4.0
Application background This paper introduces the application of artificial neural network (ANN) method in digital image recognition, and studies the existing network, and puts forward improving ideas. Key Technology Using neural network technique, a complete face recognition system is constructed. Firstly, the method of eye position estimation is used to segment the face image from the face image. Then the feature extraction is used to extract the feature. Finally, the fuzzy neural network is used to classify the images. The results show that the method is not sensitive to the pose and facial expression.
njwhat
2016-08-23
0
1
General program for the simulation of ant climbing
no vote
Application background A root of 27 cm of wood, in the first 3 cm, 7 cm, 11 cm, 17 cm, 23 cm the five positions on the ants. The wood is very thin, not through the two ant at the same time. At the beginning of the beginning of each ant's head which is not sure, they will only go forward or turn around, but will not retreat. When any two ants meet, two ants will also turn around to the opposite direction. Suppose ants can walk a centimeter per second. Write a program, the shortest time for all ants out of wood and the long time. Key Technology Solution: this may be the way most people think ofConsider the initial orientation of the ant colony, and simulate the movement of each ant to solve the problem.Solution two:Consider, although two ants meet after the turn to the opposite direction, however, can be seen as is two ants meet, pass (see here may be a lot of people have a feeling of suddenly it). That is to say that the ant movement independent,
njwhat
2016-08-23
0
1
Grid point search and simplex method Matlab source
4.0
Application background The simplex is USA mathematician G.B. Danzig first proposed in 1947. Its theory is based on: the feasible region of a linear programming problem is a n-dimensional vector space RN polyhedral convex set and its optimal value if there will be in at a vertex of the convex set. The feasible solution of the vertex is called the basic feasible solution. The basic idea of the simplex method is to find out a basic feasible solution, to identify and see if it is the optimal solution; if not, then according to certain rules to change the basic feasible solution, and then identify; if still not, then convert again, according to this repeat. Because of the limited number of basic feasible solutions, the optimal solution can be obtained by finite transformation. Key Technology Simplex search method is a direct method for unconstrained optimization. Simplex method is one of the effective methods for solving nonlinear multiple functions a
njwhat
2016-08-23
0
1
MATLAB implementation of fuzzy C means clustering
4.0
Application background Department of mathematics, graduate design, there are a lot of associated with the fuzzy mathematics. The title of my speech is on the validity of fuzzy clustering analysis, FCM algorithm is the indispensable part, here I will two iterative forms of FCM algorithm matlab code written in, perhaps some students it may come in handy Key Technology Fuzzy C means clustering (FCM), which is known as fuzzy ISODATA, is used to determine the degree of membership of each data point to a clustering algorithm. In 1973, Bezdek proposed the algorithm as an improvement of the early hard C means clustering (HCM) method.FCM n Xi (i=1,2,... , n) is divided into C fuzzy group, and each group of clustering center, so that the value function of non similarity index reached the minimum. The main difference between FCM and FCM is that the HCM uses fuzzy partition, which makes each given data point value at 0, 1 degree of membership to determine th
njwhat
2016-08-23
0
1
Image edge detection based on Ant Colony Optimizat
no vote
Application background Image edge is an important visual information, edge detection plays an important role in image processing and machine vision. Contour of the target object is detected by edge detection. The traditional edge detection method is based on the space operation, with the use of spatial differential operators to achieve, the main role of the high pass filter, such as Sobel edge operator, Roberts edge operator, Krisch edge operator, Gauss 2 Laplasse edge operator, etc. Key Technology Information hormone (Pheromone) as the basis for selecting the following behavior of ants, and through the cooperation and interaction between ants to complete the global optimization search process. The algorithm has the characteristics of strong robustness, positive feedback and distributed processing, which has been successfully used to solve many complex problems such as TSP problem, graph coloring problem and so on. But it is used in the field of
njwhat
2016-08-23
0
1
Self-learning Matlab 60 small program code
no vote
Application background MATLAB is a commercial mathematics software produced by MathWorks company, which is used in the development of algorithms, data visualization, data analysis and numerical calculation of advanced technology and interactive environment, mainly including MATLAB and Simulink two parts. Key Technology MATLAB and Maple, Mathematica are known as the three major mathematical software. It is the first in the numerical calculation of Mathematical Science and technology applications. MATLAB can perform matrix operations, drawing functions and data, the algorithm, the creation of user interface, and other programming languages, which are mainly used in engineering calculation, control design, signal processing and communication, image processing, signal detection, financial modeling design and analysis and other fields.
njwhat
2016-08-23
0
1
View More