Abstract

The issue of superresolving signal and image restoration appears in many areas, such as the restoration of diffraction-limited images, geophysical prospecting, radar and sonar signal processing, biomedical imaging, image and signal bandwidth compression, reconstruction of signals and images from partial and incomplete information, etc. There is a large repertoire of different techniques and procedures available to reconstruct such images. However, most of these techniques work well only in the absence of noise. Because this is an ill-posed problem, the actual reconstruction is sensitive to noise. Recently, new optical signal and image processing techniques have been suggested that are based on associative memory processing concepts. The purpose of this paper is to explore various associative memory processor concepts applied to the superresolving image restoration problem. Results of computer simulation will be presented.

© 1985 Optical Society of America