Share Email Print

Proceedings Paper

Optimization of computer-generated binary holograms using genetic algorithms
Author(s): Dan Cojoc; Adrian Alexandrescu
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

The aim of this paper is to compare genetic algorithms against direct point oriented coding in the design of binary phase Fourier holograms, computer generated. These are used as fan-out elements for free space optical interconnection. Genetic algorithms are optimization methods which model the natural process of genetic evolution. The configuration of the hologram is encoded to form a chromosome. To start the optimization, a population of different chromosomes randomly generated is considered. The chromosomes compete, mate and mutate until the best chromosome is obtained according to a cost function. After explaining the operators that are used by genetic algorithms, this paper presents two examples with 32 X 32 genes in a chromosome. The crossover type and the number of mutations are shown to be important factors which influence the convergence of the algorithm. GA is demonstrated to be a useful tool to design namely binary phase holograms of complicate structures.

Paper Details

Date Published: 18 November 1999
PDF: 7 pages
Proc. SPIE 3904, Fourth International Conference on Correlation Optics, (18 November 1999); doi: 10.1117/12.370408
Show Author Affiliations
Dan Cojoc, Politehnica Univ. of Bucharest (Romania)
Adrian Alexandrescu, Politehnica Univ. of Bucharest (Romania)

Published in SPIE Proceedings Vol. 3904:
Fourth International Conference on Correlation Optics
Oleg V. Angelsky, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?