Abstract

This paper explores the application of a parallel computational model, the associative network, to problems in low-level machine vision. A formal description of the associative network model is presented. Then associative networks are designed for performing Boolean functions, edge detection, and the Hough transform. Associative networks feature very flexible processor interconnections. The flexible processor interconnections allow for parallelism in the algorithm design beyond what is feasible in other parallel computational models. This work demonstrates that image processing transformations, often too slow to be practical on a sequential machine, can be executed rapidly with associative networks.

© 1987 Optical Society of America

Optical associative processor for general linear transformations

Raghuram Krishnapuram and David Casasent
Appl. Opt. 26(17) 3641-3648 (1987)

Optical implementations of associative networks with versatile adaptive learning capabilities

Arthur D. Fisher, Wendy L. Lippincott, and John N. Lee
Appl. Opt. 26(23) 5039-5054 (1987)

Massive parallelism in artificial intelligence

Lokendra Shastri
Appl. Opt. 26(10) 1829-1844 (1987)

Relational matching

Linda G. Shapiro and Robert M. Haralick
Appl. Opt. 26(10) 1845-1851 (1987)

Optoelectronic resonator neural networks

Yuri Owechko
Appl. Opt. 26(23) 5104-5111 (1987)