Share Email Print

Proceedings Paper

Neural network for optimization of binary hologram with printing model
Author(s): Guo X. Li; Reiner Eschbach; Roger L. Easton Jr.
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Hopfield neural nets are used to optimized point-oriented binary computer-generated holograms (CGH). The results are comparable to other iterative methods and may require shorter computation times. In this process, the generation of the CGH by FFT, binarization, and IFFT is viewed as a `black box' with inputs and outputs consisting of 5122 arrays containing an object of size 642. The neural-network optimization feeds back the Fourier transform of the reconstruction error to update the neuron states, which correspond to the samples of the continuous hologram. To reduce the error of the reconstruction, the input is allowed to deviate from the original array in different specified ways. For example, a previously reported approach using Projection Onto Constraint Sets varied only the region of the input array outside of the object, while we allow the entire array to be modified, thus providing more freedom in the optimization. The method may be applied either to magnitude- only or phase-only holograms. We also report on a modification of the parallel updating function. Different optimization options are compared. Use of a practical printing model requires optimization under assumed constraints to test the convergence properties of the algorithm.

Paper Details

Date Published: 12 April 1995
PDF: 14 pages
Proc. SPIE 2406, Practical Holography IX, (12 April 1995); doi: 10.1117/12.206231
Show Author Affiliations
Guo X. Li, Rochester Institute of Technology (United States)
Reiner Eschbach, Xerox Webster Research Ctr. (United States)
Roger L. Easton Jr., Rochester Institute of Technology (United States)

Published in SPIE Proceedings Vol. 2406:
Practical Holography IX
Stephen A. Benton, Editor(s)

© SPIE. Terms of Use
Back to Top