No Cover Image

Journal article

Dual-plane wavefront sensing using a vision transformer

Evan O'Rourke, Kevin O'Keeffe Orcid Logo

Optics Express, Volume: 34, Issue: 4, Pages: 6455 - 6467

Swansea University Authors: Evan O'Rourke, Kevin O'Keeffe Orcid Logo

Full text not available from this repository: check for access using links below.

Check full text

DOI (Published version): 10.1364/oe.586748

Abstract

Image-based wavefront sensing using deep-learning allows Zernike coefficients to be estimated directly from intensity measurements. To date, the majority of experiments have focused on using convolutional neural networks to estimate coefficients. Here we demonstrate a dual-plane wavefront sensor tra...

Full description

Published in: Optics Express
ISSN: 1094-4087
Published: Optica Publishing Group 2026
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa71410
first_indexed 2026-02-12T17:02:52Z
last_indexed 2026-02-13T05:35:15Z
id cronfa71410
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807 xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"><datestamp>2026-02-12T17:02:50.6155568</datestamp><bib-version>v2</bib-version><id>71410</id><entry>2026-02-12</entry><title>Dual-plane wavefront sensing using a vision transformer</title><swanseaauthors><author><sid>ef666d01c074b032cc94f53b9604eda5</sid><firstname>Evan</firstname><surname>O'Rourke</surname><name>Evan O'Rourke</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>e17dfae9042b113b28e8340ea1572db4</sid><ORCID>0000-0003-2085-0806</ORCID><firstname>Kevin</firstname><surname>O'Keeffe</surname><name>Kevin O'Keeffe</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2026-02-12</date><deptcode>BGPS</deptcode><abstract>Image-based wavefront sensing using deep-learning allows Zernike coefficients to be estimated directly from intensity measurements. To date, the majority of experiments have focused on using convolutional neural networks to estimate coefficients. Here we demonstrate a dual-plane wavefront sensor trained using a vision transformer model and compare its performance to that of the widely used convolutional neural network (CNN) architecture. Both results of experiment and simulation indicate that the vision transform can outperform the CNN where image data is significantly down sampled, due to the former's ability to more accurately predict high-order Zernike coefficients.</abstract><type>Journal Article</type><journal>Optics Express</journal><volume>34</volume><journalNumber>4</journalNumber><paginationStart>6455</paginationStart><paginationEnd>6467</paginationEnd><publisher>Optica Publishing Group</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint/><issnElectronic>1094-4087</issnElectronic><keywords/><publishedDay>23</publishedDay><publishedMonth>2</publishedMonth><publishedYear>2026</publishedYear><publishedDate>2026-02-23</publishedDate><doi>10.1364/oe.586748</doi><url>https://doi.org/10.1364/oe.586748</url><notes/><college>COLLEGE NANME</college><department>Biosciences Geography and Physics School</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>BGPS</DepartmentCode><institution>Swansea University</institution><apcterm>SU Library paid the OA fee (TA Institutional Deal)</apcterm><funders>Engineering and Physical Sciences Research Council</funders><projectreference>EP/W524694/1</projectreference><lastEdited>2026-02-12T17:02:50.6155568</lastEdited><Created>2026-02-12T16:49:19.2014716</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Biosciences, Geography and Physics - Physics</level></path><authors><author><firstname>Evan</firstname><surname>O'Rourke</surname><order>1</order></author><author><firstname>Kevin</firstname><surname>O'Keeffe</surname><orcid>0000-0003-2085-0806</orcid><order>2</order></author></authors><documents/><OutputDurs><OutputDur><Id>346</Id><DataControllerName>Kevin O'Keeffe</DataControllerName><DataControllerOrcid>0000-0003-2085-0806</DataControllerOrcid><DataControllerEmail>k.okeeffe@swansea.ac.uk</DataControllerEmail><IsDataAvailableOnline>true</IsDataAvailableOnline><DataNotAvailableOnlineReasonId xsi:nil="true"/><DurUrl>10.5281/zenodo.18268629</DurUrl><IsDurRestrictions>false</IsDurRestrictions><DurRestrictionReasonId xsi:nil="true"/><DurEmbargoDate xsi:nil="true"/></OutputDur></OutputDurs></rfc1807>
spelling 2026-02-12T17:02:50.6155568 v2 71410 2026-02-12 Dual-plane wavefront sensing using a vision transformer ef666d01c074b032cc94f53b9604eda5 Evan O'Rourke Evan O'Rourke true false e17dfae9042b113b28e8340ea1572db4 0000-0003-2085-0806 Kevin O'Keeffe Kevin O'Keeffe true false 2026-02-12 BGPS Image-based wavefront sensing using deep-learning allows Zernike coefficients to be estimated directly from intensity measurements. To date, the majority of experiments have focused on using convolutional neural networks to estimate coefficients. Here we demonstrate a dual-plane wavefront sensor trained using a vision transformer model and compare its performance to that of the widely used convolutional neural network (CNN) architecture. Both results of experiment and simulation indicate that the vision transform can outperform the CNN where image data is significantly down sampled, due to the former's ability to more accurately predict high-order Zernike coefficients. Journal Article Optics Express 34 4 6455 6467 Optica Publishing Group 1094-4087 23 2 2026 2026-02-23 10.1364/oe.586748 https://doi.org/10.1364/oe.586748 COLLEGE NANME Biosciences Geography and Physics School COLLEGE CODE BGPS Swansea University SU Library paid the OA fee (TA Institutional Deal) Engineering and Physical Sciences Research Council EP/W524694/1 2026-02-12T17:02:50.6155568 2026-02-12T16:49:19.2014716 Faculty of Science and Engineering School of Biosciences, Geography and Physics - Physics Evan O'Rourke 1 Kevin O'Keeffe 0000-0003-2085-0806 2 346 Kevin O'Keeffe 0000-0003-2085-0806 k.okeeffe@swansea.ac.uk true 10.5281/zenodo.18268629 false
title Dual-plane wavefront sensing using a vision transformer
spellingShingle Dual-plane wavefront sensing using a vision transformer
Evan O'Rourke
Kevin O'Keeffe
title_short Dual-plane wavefront sensing using a vision transformer
title_full Dual-plane wavefront sensing using a vision transformer
title_fullStr Dual-plane wavefront sensing using a vision transformer
title_full_unstemmed Dual-plane wavefront sensing using a vision transformer
title_sort Dual-plane wavefront sensing using a vision transformer
author_id_str_mv ef666d01c074b032cc94f53b9604eda5
e17dfae9042b113b28e8340ea1572db4
author_id_fullname_str_mv ef666d01c074b032cc94f53b9604eda5_***_Evan O'Rourke
e17dfae9042b113b28e8340ea1572db4_***_Kevin O'Keeffe
author Evan O'Rourke
Kevin O'Keeffe
author2 Evan O'Rourke
Kevin O'Keeffe
format Journal article
container_title Optics Express
container_volume 34
container_issue 4
container_start_page 6455
publishDate 2026
institution Swansea University
issn 1094-4087
doi_str_mv 10.1364/oe.586748
publisher Optica Publishing Group
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Biosciences, Geography and Physics - Physics{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Biosciences, Geography and Physics - Physics
url https://doi.org/10.1364/oe.586748
document_store_str 0
active_str 0
description Image-based wavefront sensing using deep-learning allows Zernike coefficients to be estimated directly from intensity measurements. To date, the majority of experiments have focused on using convolutional neural networks to estimate coefficients. Here we demonstrate a dual-plane wavefront sensor trained using a vision transformer model and compare its performance to that of the widely used convolutional neural network (CNN) architecture. Both results of experiment and simulation indicate that the vision transform can outperform the CNN where image data is significantly down sampled, due to the former's ability to more accurately predict high-order Zernike coefficients.
published_date 2026-02-23T05:35:15Z
_version_ 1856987112491974656
score 11.096068

Similar Items