Natalia A. Shevtsova1, Alain Faure2, Arkadi A. Klepatch1,
Lubov N. Podladchikova1, Ilya A. Rybak3
1 A.B.Kogan Research Institute for Neurocybernetics
Rostov State University, Rostov-on-Don, Russia
2 Laboratory LACOS, University of Le Havre, Le Havre, France
3 Central Research and Development Department
E.I. du Pont de Nemour, Wilmington, USA
A simplified neural network model of the foveal visual preprocessor (FVP) based on non-uniform representation of the vision field from its center (fovea) to the periphery has been developed. Main properties of foveal neural network and the results of computer simulation and analytical study of the model are considered. A possible implementation of foveal visual sensor on the base of the results obtained is discussed.
Keywords: neural network, vision field, fovea, periphery, visual sensor, visual preprocessor, sensory tuning.