The model of a neural network visual preprocessor

Abstract

The model of a neural network visual preprocessor and a system architecture for visual information processing are proposed. The model of the preprocessor is based on the model of a visual cortex iso-orientation domain which is considered as a neural network with retinotopically organized afferent inputs and anisotropic lateral inhibition formed by feedback connections via inhibitory interneurons. The high-level system uses the preprocessor to process image fragments with different resolutions and to represent the image as a set of contour segments of different sizes. © 1992.