Piatykop O. Processing of noisy text image on the basis of fuzzy cognitive models of visual perception

Object of research: digital image processing. The purpose of research: Development of a new information technology based on cognitive models of visual perception, to improve the efficiency of image processing text with obstacles. Methods of research: Methods of image processing, models of cognitive psychology, the apparatus of fuzzy sets, methods of programming. Theoretical and practical results: the model is developed processing and presentation of images on the levels of ganglion cells, cells of the lateral geniculate body and cells of the primary visual cortex are a universal tool that can be used in the problem of the circuit allows the presentation of the contour in the form of oriented elements, and enables more effective componentwise solution of the problem of object recognition. Image processing techniques based on cognitive models are implemented in a software package that allows you to use them in various tasks of computer vision systems. Novelty: first formalized in the form of a three-level fuzzy model of the cognitive theory of the image ganglion cells, cells of the lateral geniculate body, the cells of the primary visual cortex; improved method for processing graphic images of the text by using the proposed model, further elaborated approach illuminates in detail the character recognition, through the introduction of fuzzy models of key nodes and relations between them. А degree of implementation: modules developed software package embedded in the spam filtering information company, to develop models and methods used in the educational process at the Department of Informatics Priazovsky State Technical University in the Course "Intelligence Systems." Area of use: interbranch.