Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Powiadomienia systemowe

Sesja wygasła!

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: backpropagation learning

Sortuj według:

Ogranicz wyniki do:

Recurrent neural identification and control of a continuous bioprocess via first and second order learning

Baruch I., Mariaca-Gaspar C. R.

Journal of Automation Mobile Robotics and Intelligent Systems

2010

Vol. 4, No. 4

37-52

This paper applies a new Kalman Filter Recurrent Neural Network (KFRNN) topology and a recursive Levenberg-Mar quardt (L-M) learning algorithm capable to estimate para meters and states of highly nonlinear unknown plant in noisy environment. The proposed KFRNN identifier, learned by the Backpropagation and L-M learning algorithm, was incorporated in a direct and indirect adaptive neural con trol schemes. The proposed control schemes were applied for real-time recurrent neural identification and control of a continuous stirred tank bioreactor model, where fast convergence, noise filtering and low mean squared error of reference tracking were achieved.

Edge detection with multilayer feed forward neural network by using back propagation learning

Pahsa A.

Image Processing & Communications

2008

Vol. 13, no 1-2

21-34

In this study, availability of a multilayer feed forward neural network structure by using back propagation learning algorithm in edge detection problem is studied . Based on the edge detection problem, two types of feed forward neural network tree structures are designed. First structure is used with a 9 input for 3x3 mask and the other structure is used with 4 input for 2x2 mask. 6 black and white images are used as input to the feed forward neural network structure. 3 of the 6 images contained simple shape patterns and the rest of the 3 contained complex context. A back propagation learning technique is used in multilayer feed forward neural network that uses gradient algorithm for error minimization in hidden layers. Generated output edge profile images of the given inputs to the multilayer feed forward neural network structure are compared against the conventional edge detection techniques of Canny and Sobel mask operators.