Visual Selection and Attention Shifting Based on FitzHugh-Nagumo Equations
نویسندگان
چکیده
In this paper, we make some analysis on the FitzHugh-Nagumo model and improve it to build a neural network, and the network is used to implement visual selection and attention shifting. Each group of neurons representing one object of a visual input is synchronized; different groups of neurons representing different objects of a visual input are desynchronized. Cooperation and competition mechanism is also introduced to accelerate oscillating frequency of the salient object as well as to slow down other objects, which result in the most salient object jumping to a high frequency oscillation, while all other objects being silent. The object corresponding to high frequency oscillation is selected, then the selected object is inhibited and other neurons continue to oscillate to select the next salient object.
منابع مشابه
An improved pseudospectral approximation of generalized Burger-Huxley and Fitzhugh-Nagumo equations
In this research paper, an improved Chebyshev-Gauss-Lobatto pseudospectral approximation of nonlinear Burger-Huxley and Fitzhugh- Nagumo equations have been presented. The method employs chebyshev Gauss-Labatto points in time and space to obtain spectral accuracy. The mapping has introduced and transformed the initial-boundary value non-homogeneous problem to homogeneous problem. The main probl...
متن کاملPattern Formation of the FitzHugh-Nagumo Model: Cellular Automata Approach
FitzHugh-Nagumo (FHN) model is a famous Reaction-Diffusion System which first introduced for the conduction of electrical impulses along a nerve fiber. This model is also considered as an abstract model for pattern formation. Here, we have used the Cellular Automata method to simulate the pattern formation of the FHN model. It is shown that the pattern of this model is very similar to those...
متن کاملCanonical reduction of self-dual Yang–Mills equations to Fitzhugh–Nagumo equation and exact solutions
The (constrained) canonical reduction of four-dimensional self-dual Yang–Mills theory to two-dimensional Fitzhugh–Nagumo and the real Newell–Whitehead equations are considered. On the other hand, other methods and transformations are developed to obtain exact solutions for the original two-dimensional Fitzhugh–Nagumo and Newell– Whitehead equations. The corresponding gauge potential Al and the ...
متن کاملOptoelectronic Implementation of a FitzHugh-Nagumo Neural Model
An optoelectronic implementation of a spiking neuron model based on the FitzHugh-Nagumo equations is presented. A tunable semiconductor laser source and a spectral filter provide a nonlinear mapping from driver voltage to detected signal. Linear electronic feedback completes the implementation, which allows either electronic or optical input signals. Experimental results for a single system and...
متن کامل