Title :
Mathematical & simulation study of oscillatory neural network
Author :
Bahgat, Gehan A. ; Mahmoud, Ahmed M. ; Mashali, Samia A. ; Satama, A.E.
Author_Institution :
Dept. of Syst. & Comput., Electron. Res. Inst., Giza
Abstract :
A nonlinear oscillatory neural model is studied. It is called the modified Ellias-Grossberg model. The mathematical expression for the input range that forces it to oscillate is obtained. Simulation studies are performed to demonstrate the effect of the model parameters on the neuron behavior. Mathematical relations are deduced that can be used in tuning the model properties. The factors affecting the performance, of the interconnected neurons in a two-dimensional network, in their use in image segmentation are investigated. This paper presents the guidelines for tuning the parameters of an oscillatory neural network to meet the requirements of an application
Keywords :
image segmentation; neural nets; 2D network; coupling models; image segmentation; interconnected neuron; mathematical expression; model parameter; modified Ellias-Grossberg model; neuron behavior; nonlinear analysis; nonlinear oscillatory neural model; oscillatory neural network; Computational modeling; Computer networks; Frequency synchronization; Guidelines; Image segmentation; Mathematical model; Neural networks; Neurons; Shape; Tuning; coupling models; image segmentation; nonlinear analysis; oscillatory neural networks; parameters; synchronization;
Conference_Titel :
Circuits and Systems, 2003 IEEE 46th Midwest Symposium on
Conference_Location :
Cairo
Print_ISBN :
0-7803-8294-3
DOI :
10.1109/MWSCAS.2003.1562440
