Atmospheric modeling with the intent of training a neural net wavefront sensor

D'nardo Colucci, Michael Lloyd-Hart, Peter L. Wizinowich, J. R. Angel

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

At Steward Observatory we are developing an adaptive optics program for the Multiple Mirror Telescope (MMT) initially based in the near infrared. Using a neural network to recognize the wavefront aberrations in real time from a pair of in and out of focus images has proven itself a promising new method of wavefront sensing especially with the revolution of low noise, fast read out IR detectors. It takes a neural net on the order of 10,000 training image pairs to learn to recognize wavefront aberrations of a new, previously unseen image. Training begins with aberrated images created by the adaptive instrument itself, but since correction is over a region of approximately 2ro (Fried's parameter), the high spatial frequency components of real atmospheric turbulence are absent in these training images. We use computer simulated image pairs generated by atmospheric models based on Kolmogorov turbulence theory to further train the neural nets for the real conditions encountered when observing. Recently we have expanded our atmospheric modeling to include the stratification of turbulent layers. Using knife-edge and phase structure function measurements, we have begun to model temporal characteristics caused by atmospheric winds. The motivation for this modeling is to eventually train nets to separate the various turbulent layers allowing for multi-conjugate wavefront correction, a method which greatly extends the isoplanatic patch. Presented here are descriptions of our modeling techniques as well as results of our modeling including comparisons between stratified and single layer models.

Original languageEnglish (US)
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
PublisherPubl by Int Soc for Optical Engineering
Pages527-535
Number of pages9
Volume1688
ISBN (Print)0819408530
StatePublished - 1992
EventAtmospheric Propagation and Remote Sensing - Orlando, FL, USA
Duration: Apr 21 1992Apr 23 1992

Other

OtherAtmospheric Propagation and Remote Sensing
CityOrlando, FL, USA
Period4/21/924/23/92

Fingerprint

neural nets
Wavefronts
education
Neural networks
sensors
Sensors
Aberrations
Atmospheric turbulence
Infrared detectors
Adaptive optics
aberration
Observatories
Phase structure
Telescopes
Mirrors
Turbulence
atmospheric models
atmospheric turbulence
strata
stratification

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

Colucci, D., Lloyd-Hart, M., Wizinowich, P. L., & Angel, J. R. (1992). Atmospheric modeling with the intent of training a neural net wavefront sensor. In Proceedings of SPIE - The International Society for Optical Engineering (Vol. 1688, pp. 527-535). Publ by Int Soc for Optical Engineering.

Atmospheric modeling with the intent of training a neural net wavefront sensor. / Colucci, D'nardo; Lloyd-Hart, Michael; Wizinowich, Peter L.; Angel, J. R.

Proceedings of SPIE - The International Society for Optical Engineering. Vol. 1688 Publ by Int Soc for Optical Engineering, 1992. p. 527-535.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Colucci, D, Lloyd-Hart, M, Wizinowich, PL & Angel, JR 1992, Atmospheric modeling with the intent of training a neural net wavefront sensor. in Proceedings of SPIE - The International Society for Optical Engineering. vol. 1688, Publ by Int Soc for Optical Engineering, pp. 527-535, Atmospheric Propagation and Remote Sensing, Orlando, FL, USA, 4/21/92.
Colucci D, Lloyd-Hart M, Wizinowich PL, Angel JR. Atmospheric modeling with the intent of training a neural net wavefront sensor. In Proceedings of SPIE - The International Society for Optical Engineering. Vol. 1688. Publ by Int Soc for Optical Engineering. 1992. p. 527-535
Colucci, D'nardo ; Lloyd-Hart, Michael ; Wizinowich, Peter L. ; Angel, J. R. / Atmospheric modeling with the intent of training a neural net wavefront sensor. Proceedings of SPIE - The International Society for Optical Engineering. Vol. 1688 Publ by Int Soc for Optical Engineering, 1992. pp. 527-535
@inproceedings{f3be45efc6c244bea158075f02a3d8eb,
title = "Atmospheric modeling with the intent of training a neural net wavefront sensor",
abstract = "At Steward Observatory we are developing an adaptive optics program for the Multiple Mirror Telescope (MMT) initially based in the near infrared. Using a neural network to recognize the wavefront aberrations in real time from a pair of in and out of focus images has proven itself a promising new method of wavefront sensing especially with the revolution of low noise, fast read out IR detectors. It takes a neural net on the order of 10,000 training image pairs to learn to recognize wavefront aberrations of a new, previously unseen image. Training begins with aberrated images created by the adaptive instrument itself, but since correction is over a region of approximately 2ro (Fried's parameter), the high spatial frequency components of real atmospheric turbulence are absent in these training images. We use computer simulated image pairs generated by atmospheric models based on Kolmogorov turbulence theory to further train the neural nets for the real conditions encountered when observing. Recently we have expanded our atmospheric modeling to include the stratification of turbulent layers. Using knife-edge and phase structure function measurements, we have begun to model temporal characteristics caused by atmospheric winds. The motivation for this modeling is to eventually train nets to separate the various turbulent layers allowing for multi-conjugate wavefront correction, a method which greatly extends the isoplanatic patch. Presented here are descriptions of our modeling techniques as well as results of our modeling including comparisons between stratified and single layer models.",
author = "D'nardo Colucci and Michael Lloyd-Hart and Wizinowich, {Peter L.} and Angel, {J. R.}",
year = "1992",
language = "English (US)",
isbn = "0819408530",
volume = "1688",
pages = "527--535",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",
publisher = "Publ by Int Soc for Optical Engineering",

}

TY - GEN

T1 - Atmospheric modeling with the intent of training a neural net wavefront sensor

AU - Colucci, D'nardo

AU - Lloyd-Hart, Michael

AU - Wizinowich, Peter L.

AU - Angel, J. R.

PY - 1992

Y1 - 1992

N2 - At Steward Observatory we are developing an adaptive optics program for the Multiple Mirror Telescope (MMT) initially based in the near infrared. Using a neural network to recognize the wavefront aberrations in real time from a pair of in and out of focus images has proven itself a promising new method of wavefront sensing especially with the revolution of low noise, fast read out IR detectors. It takes a neural net on the order of 10,000 training image pairs to learn to recognize wavefront aberrations of a new, previously unseen image. Training begins with aberrated images created by the adaptive instrument itself, but since correction is over a region of approximately 2ro (Fried's parameter), the high spatial frequency components of real atmospheric turbulence are absent in these training images. We use computer simulated image pairs generated by atmospheric models based on Kolmogorov turbulence theory to further train the neural nets for the real conditions encountered when observing. Recently we have expanded our atmospheric modeling to include the stratification of turbulent layers. Using knife-edge and phase structure function measurements, we have begun to model temporal characteristics caused by atmospheric winds. The motivation for this modeling is to eventually train nets to separate the various turbulent layers allowing for multi-conjugate wavefront correction, a method which greatly extends the isoplanatic patch. Presented here are descriptions of our modeling techniques as well as results of our modeling including comparisons between stratified and single layer models.

AB - At Steward Observatory we are developing an adaptive optics program for the Multiple Mirror Telescope (MMT) initially based in the near infrared. Using a neural network to recognize the wavefront aberrations in real time from a pair of in and out of focus images has proven itself a promising new method of wavefront sensing especially with the revolution of low noise, fast read out IR detectors. It takes a neural net on the order of 10,000 training image pairs to learn to recognize wavefront aberrations of a new, previously unseen image. Training begins with aberrated images created by the adaptive instrument itself, but since correction is over a region of approximately 2ro (Fried's parameter), the high spatial frequency components of real atmospheric turbulence are absent in these training images. We use computer simulated image pairs generated by atmospheric models based on Kolmogorov turbulence theory to further train the neural nets for the real conditions encountered when observing. Recently we have expanded our atmospheric modeling to include the stratification of turbulent layers. Using knife-edge and phase structure function measurements, we have begun to model temporal characteristics caused by atmospheric winds. The motivation for this modeling is to eventually train nets to separate the various turbulent layers allowing for multi-conjugate wavefront correction, a method which greatly extends the isoplanatic patch. Presented here are descriptions of our modeling techniques as well as results of our modeling including comparisons between stratified and single layer models.

UR - http://www.scopus.com/inward/record.url?scp=0026980453&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0026980453&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0026980453

SN - 0819408530

VL - 1688

SP - 527

EP - 535

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

PB - Publ by Int Soc for Optical Engineering

ER -