Deep learning for image-based classification of OAM modes in laser beams propagating through convective turbulence

Jose Delpiano; Gustavo L. Funes; Jaime E. Cisternas; Sebastian Galaz; Jaime A. Anguita

doi:10.1117/12.2529303

Deep learning for image-based classification of OAM modes in laser beams propagating through convective turbulence

Jose Delpiano^*, Gustavo L. Funes, Jaime E. Cisternas, Sebastian Galaz, Jaime A. Anguita

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

16 Scopus citations

Abstract

Free-space optical communications are highly sensitive to distortions induced by atmospheric turbulence. This is particularly relevant when using orbital angular momentum (OAM) to send information. As current machine learning techniques for computer vision allow for accurate classification of general images, we have studied the use of a convolutional neural network for recognition of intensity patterns of OAM states after propagation experiments in a laboratory. The effect of changes in magnification and level of turbulence were explored. An error as low as 2.39% was obtained for a low level of turbulence when the training and testing data came from the same optical setup. Finally, in this article we suggest data augmentation procedures to face the problem of training before the final calibration of a communication system, with no access to data for the actual magnification and level of turbulence of real application conditions.

Original language	English
Title of host publication	Laser Communication and Propagation through the Atmosphere and Oceans VIII
Editors	Jeremy P. Bos, Alexander M. J. van Eijk, Stephen Hammel
Publisher	SPIE
ISBN (Electronic)	9781510629592
DOIs	https://doi.org/10.1117/12.2529303
State	Published - 2019
Event	Laser Communication and Propagation through the Atmosphere and Oceans VIII 2019 - San Diego, United States Duration: 12 Aug 2019 → 14 Aug 2019

Publication series

Name	Proceedings of SPIE - The International Society for Optical Engineering
Volume	11133
ISSN (Print)	0277-786X
ISSN (Electronic)	1996-756X

Conference

Conference	Laser Communication and Propagation through the Atmosphere and Oceans VIII 2019
Country/Territory	United States
City	San Diego
Period	12/08/19 → 14/08/19

Bibliographical note

Funding Information:
The authors thankfully acknowledge conversations with F. Bernuy on data augmentation techniques for deep learning. This work was partially funded by the Millennium Institute for Research in Optics (MIRO) and the Advanced Center of Electrical and Electronic Engineering (AC3E) (CONICYT/FB0008).

Publisher Copyright:
© 2019 SPIE.

Keywords

Laser beam transmission
digital image processing
neural networks
optical vortices
pattern recognition

Access to Document

10.1117/12.2529303

Cite this

Delpiano, J., Funes, G. L., Cisternas, J. E., Galaz, S., & Anguita, J. A. (2019). Deep learning for image-based classification of OAM modes in laser beams propagating through convective turbulence. In J. P. Bos, A. M. J. van Eijk, & S. Hammel (Eds.), Laser Communication and Propagation through the Atmosphere and Oceans VIII Article 1113305 (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 11133). SPIE. https://doi.org/10.1117/12.2529303

Delpiano, Jose ; Funes, Gustavo L. ; Cisternas, Jaime E. et al. / Deep learning for image-based classification of OAM modes in laser beams propagating through convective turbulence. Laser Communication and Propagation through the Atmosphere and Oceans VIII. editor / Jeremy P. Bos ; Alexander M. J. van Eijk ; Stephen Hammel. SPIE, 2019. (Proceedings of SPIE - The International Society for Optical Engineering).

@inproceedings{5c817651bb494410a7cc4b62c683c85c,

title = "Deep learning for image-based classification of OAM modes in laser beams propagating through convective turbulence",

abstract = "Free-space optical communications are highly sensitive to distortions induced by atmospheric turbulence. This is particularly relevant when using orbital angular momentum (OAM) to send information. As current machine learning techniques for computer vision allow for accurate classification of general images, we have studied the use of a convolutional neural network for recognition of intensity patterns of OAM states after propagation experiments in a laboratory. The effect of changes in magnification and level of turbulence were explored. An error as low as 2.39% was obtained for a low level of turbulence when the training and testing data came from the same optical setup. Finally, in this article we suggest data augmentation procedures to face the problem of training before the final calibration of a communication system, with no access to data for the actual magnification and level of turbulence of real application conditions.",

keywords = "Laser beam transmission, digital image processing, neural networks, optical vortices, pattern recognition",

author = "Jose Delpiano and Funes, {Gustavo L.} and Cisternas, {Jaime E.} and Sebastian Galaz and Anguita, {Jaime A.}",

note = "Funding Information: The authors thankfully acknowledge conversations with F. Bernuy on data augmentation techniques for deep learning. This work was partially funded by the Millennium Institute for Research in Optics (MIRO) and the Advanced Center of Electrical and Electronic Engineering (AC3E) (CONICYT/FB0008). Publisher Copyright: {\textcopyright} 2019 SPIE.; Laser Communication and Propagation through the Atmosphere and Oceans VIII 2019 ; Conference date: 12-08-2019 Through 14-08-2019",

year = "2019",

doi = "10.1117/12.2529303",

language = "English",

series = "Proceedings of SPIE - The International Society for Optical Engineering",

publisher = "SPIE",

editor = "Bos, {Jeremy P.} and {van Eijk}, {Alexander M. J.} and Stephen Hammel",

booktitle = "Laser Communication and Propagation through the Atmosphere and Oceans VIII",

address = "United States",

}

Delpiano, J , Funes, GL , Cisternas, JE, Galaz, S & Anguita, JA 2019, Deep learning for image-based classification of OAM modes in laser beams propagating through convective turbulence. in JP Bos, AMJ van Eijk & S Hammel (eds), Laser Communication and Propagation through the Atmosphere and Oceans VIII., 1113305, Proceedings of SPIE - The International Society for Optical Engineering, vol. 11133, SPIE, Laser Communication and Propagation through the Atmosphere and Oceans VIII 2019, San Diego, United States, 12/08/19. https://doi.org/10.1117/12.2529303

Deep learning for image-based classification of OAM modes in laser beams propagating through convective turbulence. / Delpiano, Jose ; Funes, Gustavo L.; Cisternas, Jaime E. et al.
Laser Communication and Propagation through the Atmosphere and Oceans VIII. ed. / Jeremy P. Bos; Alexander M. J. van Eijk; Stephen Hammel. SPIE, 2019. 1113305 (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 11133).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Deep learning for image-based classification of OAM modes in laser beams propagating through convective turbulence

AU - Delpiano, Jose

AU - Funes, Gustavo L.

AU - Cisternas, Jaime E.

AU - Galaz, Sebastian

AU - Anguita, Jaime A.

N1 - Funding Information: The authors thankfully acknowledge conversations with F. Bernuy on data augmentation techniques for deep learning. This work was partially funded by the Millennium Institute for Research in Optics (MIRO) and the Advanced Center of Electrical and Electronic Engineering (AC3E) (CONICYT/FB0008). Publisher Copyright: © 2019 SPIE.

PY - 2019

Y1 - 2019

N2 - Free-space optical communications are highly sensitive to distortions induced by atmospheric turbulence. This is particularly relevant when using orbital angular momentum (OAM) to send information. As current machine learning techniques for computer vision allow for accurate classification of general images, we have studied the use of a convolutional neural network for recognition of intensity patterns of OAM states after propagation experiments in a laboratory. The effect of changes in magnification and level of turbulence were explored. An error as low as 2.39% was obtained for a low level of turbulence when the training and testing data came from the same optical setup. Finally, in this article we suggest data augmentation procedures to face the problem of training before the final calibration of a communication system, with no access to data for the actual magnification and level of turbulence of real application conditions.

AB - Free-space optical communications are highly sensitive to distortions induced by atmospheric turbulence. This is particularly relevant when using orbital angular momentum (OAM) to send information. As current machine learning techniques for computer vision allow for accurate classification of general images, we have studied the use of a convolutional neural network for recognition of intensity patterns of OAM states after propagation experiments in a laboratory. The effect of changes in magnification and level of turbulence were explored. An error as low as 2.39% was obtained for a low level of turbulence when the training and testing data came from the same optical setup. Finally, in this article we suggest data augmentation procedures to face the problem of training before the final calibration of a communication system, with no access to data for the actual magnification and level of turbulence of real application conditions.

KW - Laser beam transmission

KW - digital image processing

KW - neural networks

KW - optical vortices

KW - pattern recognition

UR - http://www.scopus.com/inward/record.url?scp=85076922579&partnerID=8YFLogxK

U2 - 10.1117/12.2529303

DO - 10.1117/12.2529303

M3 - Conference contribution

AN - SCOPUS:85076922579

T3 - Proceedings of SPIE - The International Society for Optical Engineering

BT - Laser Communication and Propagation through the Atmosphere and Oceans VIII

A2 - Bos, Jeremy P.

A2 - van Eijk, Alexander M. J.

A2 - Hammel, Stephen

PB - SPIE

T2 - Laser Communication and Propagation through the Atmosphere and Oceans VIII 2019

Y2 - 12 August 2019 through 14 August 2019

ER -

Delpiano J , Funes GL , Cisternas JE, Galaz S, Anguita JA. Deep learning for image-based classification of OAM modes in laser beams propagating through convective turbulence. In Bos JP, van Eijk AMJ, Hammel S, editors, Laser Communication and Propagation through the Atmosphere and Oceans VIII. SPIE. 2019. 1113305. (Proceedings of SPIE - The International Society for Optical Engineering). doi: 10.1117/12.2529303

Deep learning for image-based classification of OAM modes in laser beams propagating through convective turbulence

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this