No Cover Image

E-Thesis 624 views 156 downloads

Classification and Segmentation of Galactic Structuresin Large Multi-spectral Images / Felix Richards

Swansea University Author: Felix Richards

  • Richards_Felix_PhD_Thesis_Final_Redacted_Signature.pdf

    PDF | E-Thesis – open access

    Copyright: The author, Felix Richards, 2023. Released under the terms of a CC-BY-NC-ND License. Third party content is excluded for use under the license terms.

    Download (60.05MB)

DOI (Published version): 10.23889/SUthesis.62601

Abstract

Extensive and exhaustive cataloguing of astronomical objects is imperative for studies seeking to understand mechanisms which drive the universe. Such cataloguing tasks can be tedious, time consuming and demand a high level of domain specific knowledge. Past astronomical imaging surveys have been ca...

Full description

Published: Swansea 2023
Institution: Swansea University
Degree level: Doctoral
Degree name: Ph.D
Supervisor: Xie, Xianghua ; Paiement, Adeline
URI: https://cronfa.swan.ac.uk/Record/cronfa62601
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2023-02-07T14:06:46Z
last_indexed 2023-02-08T04:17:09Z
id cronfa62601
recordtype RisThesis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2023-02-07T14:40:14.0466978</datestamp><bib-version>v2</bib-version><id>62601</id><entry>2023-02-07</entry><title>Classification and Segmentation of Galactic Structuresin Large Multi-spectral Images</title><swanseaauthors><author><sid>c83cb65598539b67ba65727ebb3a3f28</sid><firstname>Felix</firstname><surname>Richards</surname><name>Felix Richards</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2023-02-07</date><deptcode>SCS</deptcode><abstract>Extensive and exhaustive cataloguing of astronomical objects is imperative for studies seeking to understand mechanisms which drive the universe. Such cataloguing tasks can be tedious, time consuming and demand a high level of domain specific knowledge. Past astronomical imaging surveys have been catalogued through mostly manual effort. Immi-nent imaging surveys, however, will produce a magnitude of data that cannot be feasibly processed through manual cataloguing. Furthermore, these surveys will capture objects fainter than the night sky, termed low surface brightness objects, and at unprecedented spatial resolution owing to advancements in astronomical imaging. In this thesis, we in-vestigate the use of deep learning to automate cataloguing processes, such as detection, classification and segmentation of objects. A common theme throughout this work is the adaptation of machine learning methods to challenges specific to the domain of low surface brightness imaging.We begin with creating an annotated dataset of structures in low surface brightness images. To facilitate supervised learning in neural networks, a dataset comprised of input and corresponding ground truth target labels is required. An online tool is presented, allowing astronomers to classify and draw over objects in large multi-spectral images. A dataset produced using the tool is then detailed, containing 227 low surface brightness images from the MATLAS survey and labels made by four annotators. We then present a method for synthesising images of galactic cirrus which appear similar to MATLAS images, allowing pretraining of neural networks.A method for integrating sensitivity to orientation in convolutional neural networks is then presented. Objects in astronomical images can present in any given orientation, and thus the ability for neural networks to handle rotations is desirable. We modify con-volutional filters with sets of Gabor filters with different orientations. These orientations are learned alongside network parameters during backpropagation, allowing exact optimal orientations to be captured. The method is validated extensively on multiple datasets and use cases.We propose an attention based neural network architecture to process global contami-nants in large images. Performing analysis of low surface brightness images requires plenty of contextual information and local textual patterns. As a result, a network for processing low surface brightness images should ideally be able to accommodate large high resolu-tion images without compromising on either local or global features. We utilise attention to capture long range dependencies, and propose an efficient attention operator which significantly reduces computational cost, allowing the input of large images. We also use Gabor filters to build an attention mechanism to better capture long range orientational patterns. These techniques are validated on the task of cirrus segmentation in MAT-LAS images, and cloud segmentation on the SWIMSEG database, where state of the art performance is achieved.Following, cirrus segmentation in MATLAS images is further investigated, and a com-prehensive study is performed on the task. We discuss challenges associated with cirrus segmentation and low surface brightness images in general, and present several tech-niques to accommodate them. A novel loss function is proposed to facilitate training of the segmentation model on probabilistic targets. Results are presented on the annotated MATLAS images, with extensive ablation studies and a final benchmark to test the limits of the detailed segmentation pipeline.Finally, we develop a pipeline for multi-class segmentation of galactic structures and surrounding contaminants. Techniques of previous chapters are combined with a popu-lar instance segmentation architecture to create a neural network capable of segmenting localised objects and extended amorphous regions. The process of data preparation for training instance segmentation models is thoroughly detailed. The method is tested on segmentation of five object classes in MATLAS images. We find that unifying the tasks of galactic structure segmentation and contaminant segmentation improves model perfor-mance in comparison to isolating each task.</abstract><type>E-Thesis</type><journal/><volume/><journalNumber/><paginationStart/><paginationEnd/><publisher/><placeOfPublication>Swansea</placeOfPublication><isbnPrint/><isbnElectronic/><issnPrint/><issnElectronic/><keywords>Convolutional, Computer Vision, Artificial Intelligence, Astronomy, Segmentation, Classification</keywords><publishedDay>31</publishedDay><publishedMonth>1</publishedMonth><publishedYear>2023</publishedYear><publishedDate>2023-01-31</publishedDate><doi>10.23889/SUthesis.62601</doi><url/><notes>ORCiD identifier: https://orcid.org/0000-0001-8558-9691</notes><college>COLLEGE NANME</college><department>Computer Science</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>SCS</DepartmentCode><institution>Swansea University</institution><supervisor>Xie, Xianghua ; Paiement, Adeline</supervisor><degreelevel>Doctoral</degreelevel><degreename>Ph.D</degreename><degreesponsorsfunders>STFC (ST/P006779/1)</degreesponsorsfunders><apcterm/><funders/><projectreference/><lastEdited>2023-02-07T14:40:14.0466978</lastEdited><Created>2023-02-07T14:02:18.8014185</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Mathematics and Computer Science - Computer Science</level></path><authors><author><firstname>Felix</firstname><surname>Richards</surname><order>1</order></author></authors><documents><document><filename>62601__26506__0e84e669f82846efa65738ef0ca8b36c.pdf</filename><originalFilename>Richards_Felix_PhD_Thesis_Final_Redacted_Signature.pdf</originalFilename><uploaded>2023-02-07T14:16:03.0239567</uploaded><type>Output</type><contentLength>62968022</contentLength><contentType>application/pdf</contentType><version>E-Thesis &#x2013; open access</version><cronfaStatus>true</cronfaStatus><documentNotes>Copyright: The author, Felix Richards, 2023. Released under the terms of a CC-BY-NC-ND License. Third party content is excluded for use under the license terms.</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>https://creativecommons.org/licenses/by-nc-nd/4.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling 2023-02-07T14:40:14.0466978 v2 62601 2023-02-07 Classification and Segmentation of Galactic Structuresin Large Multi-spectral Images c83cb65598539b67ba65727ebb3a3f28 Felix Richards Felix Richards true false 2023-02-07 SCS Extensive and exhaustive cataloguing of astronomical objects is imperative for studies seeking to understand mechanisms which drive the universe. Such cataloguing tasks can be tedious, time consuming and demand a high level of domain specific knowledge. Past astronomical imaging surveys have been catalogued through mostly manual effort. Immi-nent imaging surveys, however, will produce a magnitude of data that cannot be feasibly processed through manual cataloguing. Furthermore, these surveys will capture objects fainter than the night sky, termed low surface brightness objects, and at unprecedented spatial resolution owing to advancements in astronomical imaging. In this thesis, we in-vestigate the use of deep learning to automate cataloguing processes, such as detection, classification and segmentation of objects. A common theme throughout this work is the adaptation of machine learning methods to challenges specific to the domain of low surface brightness imaging.We begin with creating an annotated dataset of structures in low surface brightness images. To facilitate supervised learning in neural networks, a dataset comprised of input and corresponding ground truth target labels is required. An online tool is presented, allowing astronomers to classify and draw over objects in large multi-spectral images. A dataset produced using the tool is then detailed, containing 227 low surface brightness images from the MATLAS survey and labels made by four annotators. We then present a method for synthesising images of galactic cirrus which appear similar to MATLAS images, allowing pretraining of neural networks.A method for integrating sensitivity to orientation in convolutional neural networks is then presented. Objects in astronomical images can present in any given orientation, and thus the ability for neural networks to handle rotations is desirable. We modify con-volutional filters with sets of Gabor filters with different orientations. These orientations are learned alongside network parameters during backpropagation, allowing exact optimal orientations to be captured. The method is validated extensively on multiple datasets and use cases.We propose an attention based neural network architecture to process global contami-nants in large images. Performing analysis of low surface brightness images requires plenty of contextual information and local textual patterns. As a result, a network for processing low surface brightness images should ideally be able to accommodate large high resolu-tion images without compromising on either local or global features. We utilise attention to capture long range dependencies, and propose an efficient attention operator which significantly reduces computational cost, allowing the input of large images. We also use Gabor filters to build an attention mechanism to better capture long range orientational patterns. These techniques are validated on the task of cirrus segmentation in MAT-LAS images, and cloud segmentation on the SWIMSEG database, where state of the art performance is achieved.Following, cirrus segmentation in MATLAS images is further investigated, and a com-prehensive study is performed on the task. We discuss challenges associated with cirrus segmentation and low surface brightness images in general, and present several tech-niques to accommodate them. A novel loss function is proposed to facilitate training of the segmentation model on probabilistic targets. Results are presented on the annotated MATLAS images, with extensive ablation studies and a final benchmark to test the limits of the detailed segmentation pipeline.Finally, we develop a pipeline for multi-class segmentation of galactic structures and surrounding contaminants. Techniques of previous chapters are combined with a popu-lar instance segmentation architecture to create a neural network capable of segmenting localised objects and extended amorphous regions. The process of data preparation for training instance segmentation models is thoroughly detailed. The method is tested on segmentation of five object classes in MATLAS images. We find that unifying the tasks of galactic structure segmentation and contaminant segmentation improves model perfor-mance in comparison to isolating each task. E-Thesis Swansea Convolutional, Computer Vision, Artificial Intelligence, Astronomy, Segmentation, Classification 31 1 2023 2023-01-31 10.23889/SUthesis.62601 ORCiD identifier: https://orcid.org/0000-0001-8558-9691 COLLEGE NANME Computer Science COLLEGE CODE SCS Swansea University Xie, Xianghua ; Paiement, Adeline Doctoral Ph.D STFC (ST/P006779/1) 2023-02-07T14:40:14.0466978 2023-02-07T14:02:18.8014185 Faculty of Science and Engineering School of Mathematics and Computer Science - Computer Science Felix Richards 1 62601__26506__0e84e669f82846efa65738ef0ca8b36c.pdf Richards_Felix_PhD_Thesis_Final_Redacted_Signature.pdf 2023-02-07T14:16:03.0239567 Output 62968022 application/pdf E-Thesis – open access true Copyright: The author, Felix Richards, 2023. Released under the terms of a CC-BY-NC-ND License. Third party content is excluded for use under the license terms. true eng https://creativecommons.org/licenses/by-nc-nd/4.0/
title Classification and Segmentation of Galactic Structuresin Large Multi-spectral Images
spellingShingle Classification and Segmentation of Galactic Structuresin Large Multi-spectral Images
Felix Richards
title_short Classification and Segmentation of Galactic Structuresin Large Multi-spectral Images
title_full Classification and Segmentation of Galactic Structuresin Large Multi-spectral Images
title_fullStr Classification and Segmentation of Galactic Structuresin Large Multi-spectral Images
title_full_unstemmed Classification and Segmentation of Galactic Structuresin Large Multi-spectral Images
title_sort Classification and Segmentation of Galactic Structuresin Large Multi-spectral Images
author_id_str_mv c83cb65598539b67ba65727ebb3a3f28
author_id_fullname_str_mv c83cb65598539b67ba65727ebb3a3f28_***_Felix Richards
author Felix Richards
author2 Felix Richards
format E-Thesis
publishDate 2023
institution Swansea University
doi_str_mv 10.23889/SUthesis.62601
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 Mathematics and Computer Science - Computer Science{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Mathematics and Computer Science - Computer Science
document_store_str 1
active_str 0
description Extensive and exhaustive cataloguing of astronomical objects is imperative for studies seeking to understand mechanisms which drive the universe. Such cataloguing tasks can be tedious, time consuming and demand a high level of domain specific knowledge. Past astronomical imaging surveys have been catalogued through mostly manual effort. Immi-nent imaging surveys, however, will produce a magnitude of data that cannot be feasibly processed through manual cataloguing. Furthermore, these surveys will capture objects fainter than the night sky, termed low surface brightness objects, and at unprecedented spatial resolution owing to advancements in astronomical imaging. In this thesis, we in-vestigate the use of deep learning to automate cataloguing processes, such as detection, classification and segmentation of objects. A common theme throughout this work is the adaptation of machine learning methods to challenges specific to the domain of low surface brightness imaging.We begin with creating an annotated dataset of structures in low surface brightness images. To facilitate supervised learning in neural networks, a dataset comprised of input and corresponding ground truth target labels is required. An online tool is presented, allowing astronomers to classify and draw over objects in large multi-spectral images. A dataset produced using the tool is then detailed, containing 227 low surface brightness images from the MATLAS survey and labels made by four annotators. We then present a method for synthesising images of galactic cirrus which appear similar to MATLAS images, allowing pretraining of neural networks.A method for integrating sensitivity to orientation in convolutional neural networks is then presented. Objects in astronomical images can present in any given orientation, and thus the ability for neural networks to handle rotations is desirable. We modify con-volutional filters with sets of Gabor filters with different orientations. These orientations are learned alongside network parameters during backpropagation, allowing exact optimal orientations to be captured. The method is validated extensively on multiple datasets and use cases.We propose an attention based neural network architecture to process global contami-nants in large images. Performing analysis of low surface brightness images requires plenty of contextual information and local textual patterns. As a result, a network for processing low surface brightness images should ideally be able to accommodate large high resolu-tion images without compromising on either local or global features. We utilise attention to capture long range dependencies, and propose an efficient attention operator which significantly reduces computational cost, allowing the input of large images. We also use Gabor filters to build an attention mechanism to better capture long range orientational patterns. These techniques are validated on the task of cirrus segmentation in MAT-LAS images, and cloud segmentation on the SWIMSEG database, where state of the art performance is achieved.Following, cirrus segmentation in MATLAS images is further investigated, and a com-prehensive study is performed on the task. We discuss challenges associated with cirrus segmentation and low surface brightness images in general, and present several tech-niques to accommodate them. A novel loss function is proposed to facilitate training of the segmentation model on probabilistic targets. Results are presented on the annotated MATLAS images, with extensive ablation studies and a final benchmark to test the limits of the detailed segmentation pipeline.Finally, we develop a pipeline for multi-class segmentation of galactic structures and surrounding contaminants. Techniques of previous chapters are combined with a popu-lar instance segmentation architecture to create a neural network capable of segmenting localised objects and extended amorphous regions. The process of data preparation for training instance segmentation models is thoroughly detailed. The method is tested on segmentation of five object classes in MATLAS images. We find that unifying the tasks of galactic structure segmentation and contaminant segmentation improves model perfor-mance in comparison to isolating each task.
published_date 2023-01-31T04:22:20Z
_version_ 1763754472439283712
score 11.035655

Similar Items