Share Email Print

Proceedings Paper

Inverse and direct problems of optics: usage of artificial neural networks
Author(s): Victor Sergeyevich Abrukov; Roman Ivanovich Pavlov; Gennadiy I. Malinin
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

We describe an application of artificial neural networks (ANN) for solving of inverse and direct problems of optics. Using the ANN we calculate local and integral characteristics of object by means of incomplete set of data that characterize optical images. Possibilities of usage the only one value of a function of signal intensity distributionn in a plane of a registration for full determination of distribution of local characteristics in an object are shown. It is very important for optical fiber sensors, smart sensors and MEMS. Examples of ANN usage for a case of object with a cylindrical symmetry in a field of interferometry are presented. Results obtained show that determination of object local and integral characteristics can be perform very much simpler than by means of standard procedures and numerical approaches for signal processing, reduction and analysis. The ANN can allow also to solve number of tasks that could not be solved by means of usual approaches. In prospects, this method can be used for creation of automated systems for diagnostics, testing and control in various fields of scientific and applied research as well as in industry.

Paper Details

Date Published: 2 August 2004
PDF: 4 pages
Proc. SPIE 5531, Interferometry XII: Techniques and Analysis, (2 August 2004); doi: 10.1117/12.562516
Show Author Affiliations
Victor Sergeyevich Abrukov, Chuvash State Univ. (Russia)
Moscow State Open Univ. (Russia)
Roman Ivanovich Pavlov, Chuvash State Univ. (Russia)
Moscow State Open Univ. (Russia)
Gennadiy I. Malinin, Chuvash State Univ. (Russia)
Moscow State Open Univ. (Russia)

Published in SPIE Proceedings Vol. 5531:
Interferometry XII: Techniques and Analysis
Katherine Creath; Joanna Schmit, Editor(s)

© SPIE. Terms of Use
Back to Top