recurrent neural network r

... candidate text regions. Those features are used as the input of a neural network for training based on the back-propagation algorithm for neural networks. After the neural network is well trained, ... on a neural network. The training of the neural network is based on the features we obtain from the DWT detail component sub-bands. As shown in Figure 6, the proposed neural network architecture ... a method for extracting text regions from static images or video sequences using DWT and a neural network. DWT provides features of text regions for the training of a neural network using...

Ngày tải lên: 05/11/2012, 14:51

... average color and colors larger than the average color. Basically these features represent the central positions (average color) and the standard errors of three parts: the whole histogram, ... having color smaller than the average and pixels having color larger than the average accordingly. The next three are the standard errors of number of pixels for all colors; colors smaller than ... Classification result Figure 9. Binarization result IV. CONCLUSION AND FUTURE WORKS A neural network based method is discussed in this paper. The features used for the neural network are not only...

Ngày tải lên: 05/11/2012, 14:54

Ngày tải lên: 05/09/2013, 14:58

Ngày tải lên: 12/09/2013, 16:20

... Tính toán lỗi (error), điểm ra (output) và trọng số (weight) thường xuyên. - Xây dựng mạng Neural Network. 1) Huấn luyện mạng Lưu đồ giải thuật huấn luyện mạng của chương trình: 2) Thực ... [3] Artificial Intelligence and cognitive science. © 2006, Nils J.Nilsson Stanford AI Lab http://ai.stanford.edu/~nilsson [4] Offline Handwring Recognition Using Artificial Neural Networks ... và tương ứng với những kí tự đầu ra tương ứng trong một files riêng biệt cho tập hợp những kí tự huấn luyện (*.cts – character trainer sets). - Tải và lưu trữ thường xuyên mạng đã huấn luyện....

Ngày tải lên: 06/10/2013, 20:20

... nonlinear dynamic system theory that incorporates neural networks is still an ongoing research area. 1.3.1.2 Nonlinear Filtering There are many reports on using artificial neural networks to perform ... using neural networks. In the first part, in-depth surveys of recent progress of neural network computing paradigms are presented. Part One consists of five chapters: • Chapter 1: Introduction to Neural ... different kinds of nonlinear networks with the same theoretical pattern storage (or pattern memorization) produce very similar values of the training error E [39, 40]. In order to verify the observation...

Ngày tải lên: 12/12/2013, 23:15

... that progress has a direct bearing on signal processing. In particular, the nonlinear nature of neural networks, the ability of neural networks to learn from their environments in super- vised ... square errors for l=2:L, % the l-th layer u{l}=w{l}*[ones(1,K);z{l-1}]; % u{l} is n(l) by K z{l}=actfun(u{l},atype(l)); end error=d-z{L}; % error is N by K E(t)=sum(sum(error.*error)); % Error ... powerful modeling tools motivated numerous research efforts to explore new signal processing applications of artificial neural networks. During the course of the research, many neural network paradigms...

Ngày tải lên: 12/12/2013, 23:15

... processing information in a parallel distributed fashion. Feedforward neural networks and recur- rent neural networks are two major classes of artificial neural networks. Feedforward neural networks, ... process control. The neural network models used were multilayer feedforward networks, ART networks, and cerebellar model artic- ulation controller, as shown in Figure 2.4. Neural networks are promising ... feedforward neural network trained using the BP algorithm to control a turning process. Given feed rate error and change in the error, the trained neural network was able to recommend the control...

Ngày tải lên: 17/12/2013, 06:15

... block to generate the PWM outputs. - for for for for for for for for for for (5) - for for for for for for for for for for (6) 664 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 38, NO. 3, MAY/JUNE ... two-level inverter and, therefore, the SVM algorithm to be implemented in a neural network is considerably more complex. In the proposed scheme, a three-layer feedforward neural network receives the command ... Fig. 4 for all the regions in the six sectors. For example, the phase- turn-on time expressions in mode 1 can be derived as - for for for for for for (3) - for for for for for for (4) where and...

Ngày tải lên: 22/12/2013, 08:16

Ngày tải lên: 29/12/2013, 11:17

... symmetric cryptography based on chaotic signal generator and a clipped neural network. Advances in Neural Networks-ISNN, Intl. Symp. Neural Networks Proc., Part II. Lecture Notes in Computer Science, ... overall error to test network performance. The training set is repeatedly presented to the network and the weight values are adjusted until the overall error is below a predetermined tolerance. ... reasonable outputs will result before training. The weights are adjusted to reduce the error by propagating the output error backward through the network. This process is where the backpropagation...

Ngày tải lên: 14/02/2014, 16:20

... between predictors (such as meteorological and air quality variables) and predictand (in this case, hourly concentration). Neural networks, in particular the multi-layer perceptron (Hornik et ... Press, Ann Arbor. Hornik, K., Stinchcombe, M., White, H., 1989. Multilayer feedfor- ward networks are universal approximators. Neural Networks 2, 359–366. Jain, A., Zongker, D., 1997. Feature selection: ... The parameters used were 25 hidden neurons, learning algorithm of scaled conjugated gradient back-propagation, the per- formance function of regularized mean squared error (RMSE), hyperbolic...

Ngày tải lên: 17/02/2014, 22:20

Ngày tải lên: 19/02/2014, 09:20

Ngày tải lên: 17/03/2014, 04:20