Share Email Print

Proceedings Paper

Foreground/background segmentation of optical character recognition (OCR) labels by a single-layer recurrent neural network
Author(s): Lee F. Holeva
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

This paper describes the development of a recurrent neural network to segment gray scale label images into binary label images. To determine a pixel label, the neural network takes into account three sources of information: pixel intensities, correlations between neighboring labels, and edge gradients. These three sources of information are succinctly combined via the network's energy function. By changing its label state to minimize the energy function, the network satisfies constraints imposed by the input image and the current label values. To be mappable to analog hardware, it is desirable that the neural equations be deterministic. Two deterministic networks are developed and compared. The first operates at the zero temperature limit, the original Hopfield network. The second employs the mean field annealing algorithm. It is shown that with only a moderate increase in computational requirements, the mean field approach produces far superior results.

Paper Details

Date Published: 6 April 1995
PDF: 13 pages
Proc. SPIE 2492, Applications and Science of Artificial Neural Networks, (6 April 1995); doi: 10.1117/12.205182
Show Author Affiliations
Lee F. Holeva, United Parcel Service Research & Development (United States)

Published in SPIE Proceedings Vol. 2492:
Applications and Science of Artificial Neural Networks
Steven K. Rogers; Dennis W. Ruck, Editor(s)

© SPIE. Terms of Use
Back to Top