Optical vortices localization with artificial neural network

Wojciech Fraczek; Tomasz Guszkowski

doi:10.1117/12.881585

14 December 2010 Optical vortices localization with artificial neural network

Wojciech Fraczek, Tomasz Guszkowski

Proceedings Volume 7746, 17th Slovak-Czech-Polish Optical Conference on Wave and Quantum Aspects of Contemporary Optics; 77460C (2010) https://doi.org/10.1117/12.881585
Event: 17th Slovak-Czech-Polish Optical Conference on Wave and Quantum Aspects of Contemporary Optics, 2010, Liptovsky Jan, Slovakia

Abstract

This paper presents a new technique of localization of Optical Vortex center (Vortex Point), which is based on Artificial Neural Network. The network accepts as inputs intensities of pixels in a square sub-area of fixed size and produces two output signals correlated to X and Y coordinates of vortex point. A feed-forward neural network with one hidden, non-linear layer was used. The learning criterion was Mean Squared Error and the net was trained with Levenberg-Marquardt algorithm implemented in MATLAB with distorted and noisy images of simulated Optical Vortex Interferometer. The authors provide the results of localization for two sizes of sub-area and two types of simulated interference patterns. The first type of interferogram contain lattice of optical vortices obtained by interference of three plane waves and the second type of interferogram with fringe bifurcations created by interference of lattice of optical vortex with additional, fourth plane wave. The results of simulations suggest possible employment of the proposed method in an experimental setup due to the localization error median being around 0.4 pixel for "three waves" case and 0.9 pixel for "four waves".

Citation Download Citation

Wojciech Fraczek and Tomasz Guszkowski "Optical vortices localization with artificial neural network", Proc. SPIE 7746, 17th Slovak-Czech-Polish Optical Conference on Wave and Quantum Aspects of Contemporary Optics, 77460C (14 December 2010); https://doi.org/10.1117/12.881585

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available