Recognition of visual patterns based on topological information processing

  • V. V. Demyokhin Donetsk National University, Ukraine
  • V. V. Danilov Donetsk National University, Ukraine
Keywords: pattern classification, stochastic connections, associative elements

Abstract

A method and device for image classification have been developed, enabling determination of the class membership of visual patterns. The method is based on changes in the topology of stochastic connections between sensors and associative elements (A-elements), which depend on the distribution of brightness levels at the input of the recognition system and the distribution of reflective (transmissive) states of the A-element matrix established during training.

Published
2007-06-29
How to Cite
Demyokhin, V. V., & Danilov, V. V. (2007). Recognition of visual patterns based on topological information processing. Technology and Design in Electronic Equipment, (3), 24-29. Retrieved from https://www.tkea.com.ua/index.php/journal/article/view/TKEA2007.3.24
Issue
No. 3 (2007): Tekhnologiya i konstruirovanie v elektronnoi apparature
Section
Articles

Copyright (c) 2007 Demyokhin V. V., Danilov V. V.

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.