Share Email Print

Proceedings Paper

Representing Shape Primitives In Neural Networks
Author(s): Ted Pawlicki
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Parallel distributed, connectionist, neural networks present powerful computational metaphors for diverse applications ranging from machine perception to artificial intelligence [1-3,6]. Historically, such systems have been appealing for their ability to perform self-organization and learning[7, 8, 11]. However, while simple systems of this type can perform interesting tasks, results from such systems perform little better than existing template matchers in some real world applications [9,10]. The definition of a more complex structure made from simple units can be used to enhance performance of these models [4, 5], but the addition of extra complexity raises representational issues. This paper reports on attempts to code information and features which have classically been useful to shape analysis into a neural network system.

Paper Details

Date Published: 22 August 1988
PDF: 7 pages
Proc. SPIE 0938, Digital and Optical Shape Representation and Pattern Recognition, (22 August 1988); doi: 10.1117/12.976624
Show Author Affiliations
Ted Pawlicki, State University of New York at Buffalo (United States)

Published in SPIE Proceedings Vol. 0938:
Digital and Optical Shape Representation and Pattern Recognition
Richard D. Juday, Editor(s)

© SPIE. Terms of Use
Back to Top