A neural architecture applied to the enhancement of noisy binary images

Document Type

Article

Publication Date

1-1-1992

Abstract

Neural networks represent a new dimension in computing, both in concept and in implementation. Their applications are varied with many significant and exciting developments which have been developed in the field of image processing, among others. This paper presents the formulation of one improved architecture, a modified Adaptive Resonance Theory (ART), for the enhancement of noisy binary images. The operation and performance of the traditional ART1 in classifying binary input patterns are first investigated. Based upon the ART1, a noise-filtering architecture is devised, whereby pre-established recognition categories are used as edge-detection exemplars. Experimental results of applying the architecture to perform edge detection and region enhancement are also provided. © 1992.

Identifier

33746146538 (Scopus)

Publication Title

Engineering Applications of Artificial Intelligence

External Full Text Location

https://doi.org/10.1016/0952-1976(92)90005-5

ISSN

09521976

First Page

215

Last Page

222

Issue

3

Volume

5

This document is currently not available here.

Share

COinS