No Cover Image

Conference Paper/Proceeding/Abstract 25 views

SIGNN - Star Identification using Graph Neural Networks

Floyd Hepburn-Dickins, Mark Jones Orcid Logo, Mike Edwards Orcid Logo, Jay Morgan Orcid Logo, Steve Bell

Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), Feb. 2025

Swansea University Authors: Floyd Hepburn-Dickins, Mark Jones Orcid Logo, Mike Edwards Orcid Logo, Jay Morgan Orcid Logo

Abstract

As a solution for the lost-in-space star identification problem we present Star Identification using Graph Neural Network (SIGNN), a novel approach using Graph Attention Networks. By representing the celestial sphere as a graph data structure, created from the ESA's Hipparcos catalogue, we are...

Full description

Published in: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), Feb. 2025
Published:
URI: https://cronfa.swan.ac.uk/Record/cronfa68091
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2024-10-29T11:03:15Z
last_indexed 2024-10-29T11:03:15Z
id cronfa68091
recordtype SURis
fullrecord <?xml version="1.0" encoding="utf-8"?><rfc1807 xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xsd="http://www.w3.org/2001/XMLSchema"><bib-version>v2</bib-version><id>68091</id><entry>2024-10-29</entry><title>SIGNN - Star Identification using Graph Neural Networks</title><swanseaauthors><author><sid>d8ecf05934e394b7bd020a2ce2c11d0c</sid><firstname>Floyd</firstname><surname>Hepburn-Dickins</surname><name>Floyd Hepburn-Dickins</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>2e1030b6e14fc9debd5d5ae7cc335562</sid><ORCID>0000-0001-8991-1190</ORCID><firstname>Mark</firstname><surname>Jones</surname><name>Mark Jones</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>684864a1ce01c3d774e83ed55e41770e</sid><ORCID>0000-0003-3367-969X</ORCID><firstname>Mike</firstname><surname>Edwards</surname><name>Mike Edwards</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>df9a27bcf77b4769c2ebbb702b587491</sid><ORCID>0000-0003-3719-362X</ORCID><firstname>Jay</firstname><surname>Morgan</surname><name>Jay Morgan</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2024-10-29</date><abstract>As a solution for the lost-in-space star identification problem we present Star Identification using Graph Neural Network (SIGNN), a novel approach using Graph Attention Networks. By representing the celestial sphere as a graph data structure, created from the ESA's Hipparcos catalogue, we are able to accurately capture the rich information and relationships within local star fields. Graph learning techniques allow our model to aggregate information and learn the relative importance of the nodes and structure within each stars local neighbourhood to it's identification. This approach, combined with our parametric data-generation and noise simulation, allows us to train a highly robust model capable of accurate star identification even under intensive noise, outperforming existing methods. Code and generation techniques will be available on github.com.</abstract><type>Conference Paper/Proceeding/Abstract</type><journal>Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), Feb. 2025</journal><volume/><journalNumber/><paginationStart/><paginationEnd/><publisher/><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint/><issnElectronic/><keywords/><publishedDay>0</publishedDay><publishedMonth>0</publishedMonth><publishedYear>0</publishedYear><publishedDate>0001-01-01</publishedDate><doi/><url/><notes/><college>COLLEGE NANME</college><CollegeCode>COLLEGE CODE</CollegeCode><institution>Swansea University</institution><apcterm>Not Required</apcterm><funders>EPSRC</funders><projectreference>EP/S021892/1</projectreference><lastEdited>2024-10-29T11:12:12.2189113</lastEdited><Created>2024-10-29T10:58:16.3133961</Created><path><level id="1">College of Science</level><level id="2">Computer Science</level></path><authors><author><firstname>Floyd</firstname><surname>Hepburn-Dickins</surname><order>1</order></author><author><firstname>Mark</firstname><surname>Jones</surname><orcid>0000-0001-8991-1190</orcid><order>2</order></author><author><firstname>Mike</firstname><surname>Edwards</surname><orcid>0000-0003-3367-969X</orcid><order>3</order></author><author><firstname>Jay</firstname><surname>Morgan</surname><orcid>0000-0003-3719-362X</orcid><order>4</order></author><author><firstname>Steve</firstname><surname>Bell</surname><order>5</order></author></authors><documents/><OutputDurs/></rfc1807>
spelling v2 68091 2024-10-29 SIGNN - Star Identification using Graph Neural Networks d8ecf05934e394b7bd020a2ce2c11d0c Floyd Hepburn-Dickins Floyd Hepburn-Dickins true false 2e1030b6e14fc9debd5d5ae7cc335562 0000-0001-8991-1190 Mark Jones Mark Jones true false 684864a1ce01c3d774e83ed55e41770e 0000-0003-3367-969X Mike Edwards Mike Edwards true false df9a27bcf77b4769c2ebbb702b587491 0000-0003-3719-362X Jay Morgan Jay Morgan true false 2024-10-29 As a solution for the lost-in-space star identification problem we present Star Identification using Graph Neural Network (SIGNN), a novel approach using Graph Attention Networks. By representing the celestial sphere as a graph data structure, created from the ESA's Hipparcos catalogue, we are able to accurately capture the rich information and relationships within local star fields. Graph learning techniques allow our model to aggregate information and learn the relative importance of the nodes and structure within each stars local neighbourhood to it's identification. This approach, combined with our parametric data-generation and noise simulation, allows us to train a highly robust model capable of accurate star identification even under intensive noise, outperforming existing methods. Code and generation techniques will be available on github.com. Conference Paper/Proceeding/Abstract Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), Feb. 2025 0 0 0 0001-01-01 COLLEGE NANME COLLEGE CODE Swansea University Not Required EPSRC EP/S021892/1 2024-10-29T11:12:12.2189113 2024-10-29T10:58:16.3133961 College of Science Computer Science Floyd Hepburn-Dickins 1 Mark Jones 0000-0001-8991-1190 2 Mike Edwards 0000-0003-3367-969X 3 Jay Morgan 0000-0003-3719-362X 4 Steve Bell 5
title SIGNN - Star Identification using Graph Neural Networks
spellingShingle SIGNN - Star Identification using Graph Neural Networks
Floyd Hepburn-Dickins
Mark Jones
Mike Edwards
Jay Morgan
title_short SIGNN - Star Identification using Graph Neural Networks
title_full SIGNN - Star Identification using Graph Neural Networks
title_fullStr SIGNN - Star Identification using Graph Neural Networks
title_full_unstemmed SIGNN - Star Identification using Graph Neural Networks
title_sort SIGNN - Star Identification using Graph Neural Networks
author_id_str_mv d8ecf05934e394b7bd020a2ce2c11d0c
2e1030b6e14fc9debd5d5ae7cc335562
684864a1ce01c3d774e83ed55e41770e
df9a27bcf77b4769c2ebbb702b587491
author_id_fullname_str_mv d8ecf05934e394b7bd020a2ce2c11d0c_***_Floyd Hepburn-Dickins
2e1030b6e14fc9debd5d5ae7cc335562_***_Mark Jones
684864a1ce01c3d774e83ed55e41770e_***_Mike Edwards
df9a27bcf77b4769c2ebbb702b587491_***_Jay Morgan
author Floyd Hepburn-Dickins
Mark Jones
Mike Edwards
Jay Morgan
author2 Floyd Hepburn-Dickins
Mark Jones
Mike Edwards
Jay Morgan
Steve Bell
format Conference Paper/Proceeding/Abstract
container_title Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), Feb. 2025
institution Swansea University
college_str College of Science
hierarchytype
hierarchy_top_id collegeofscience
hierarchy_top_title College of Science
hierarchy_parent_id collegeofscience
hierarchy_parent_title College of Science
department_str Computer Science{{{_:::_}}}College of Science{{{_:::_}}}Computer Science
document_store_str 0
active_str 0
description As a solution for the lost-in-space star identification problem we present Star Identification using Graph Neural Network (SIGNN), a novel approach using Graph Attention Networks. By representing the celestial sphere as a graph data structure, created from the ESA's Hipparcos catalogue, we are able to accurately capture the rich information and relationships within local star fields. Graph learning techniques allow our model to aggregate information and learn the relative importance of the nodes and structure within each stars local neighbourhood to it's identification. This approach, combined with our parametric data-generation and noise simulation, allows us to train a highly robust model capable of accurate star identification even under intensive noise, outperforming existing methods. Code and generation techniques will be available on github.com.
published_date 0001-01-01T11:12:10Z
_version_ 1814246542192672768
score 11.0351515