Morphological evolution of galaxies with deep learning

Mitchell Cavanagh

doi:10.26182/ckrn-0t69

Morphological evolution of galaxies with deep learning

Mitchell Cavanagh

Research output: Thesis › Doctoral Thesis

238 Downloads (Pure)

Abstract

Understanding the changes in galaxy morphology over time can yield important insights into the physics behind galaxy evolution. In this thesis, I adopt a deep learning approach to morphological classification. I develop a new suite of CNNs to rapidly and accurately classify galaxy morphologies. I apply these models to classify tens of thousands of galaxies in several datasets across observational surveys and simulations, analysing the evolution of barred galaxies in EAGLE, the growth of lenticular galaxies over time in COSMOS, and the kinematics of different morphologies in SAMI. I also develop an image segmentation model capable of analysing galaxy bars.

Original language	English
Qualification	Doctor of Philosophy
Awarding Institution	The University of Western Australia
Supervisors/Advisors	Bekki, Kenji, Supervisor Groves, Brent, Supervisor
Thesis sponsors	Australian Government Research Training Program
Award date	6 Jun 2023
DOIs	https://doi.org/10.26182/ckrn-0t69
Publication status	Unpublished - 2023

Access to Document

10.26182/ckrn-0t69

THESIS - DOCTOR OF PHILOSOPHY - CAVANAGH Mitchell - 2023
This work is protected by Copyright. You may print or download ONE copy of this document for the purpose of your own non-commercial research or study. Any other use requires permission from the copyright owner. The Copyright Act requires you to attribute any copyright works you quote or paraphrase.

3 Article

The redshift evolution of the S0 fraction for z < 1 in COSMOS
Cavanagh, M., Bekki, K. & Groves, B., 1 Apr 2023, In: Monthly Notices of the Royal Astronomical Society. 520, 4, p. 5885-5902 18 p.
Research output: Contribution to journal › Article › peer-review

Open Access
File
5 Citations (Scopus)

44 Downloads (Pure)
The evolution of barred galaxies in the EAGLE simulations
Cavanagh, M. K., Bekki, K., Groves, B. A. & Pfeffer, J., 1 Mar 2022, In: Monthly Notices of the Royal Astronomical Society. 510, 4, p. 5164-5178 15 p.
Research output: Contribution to journal › Article › peer-review
18 Citations (Scopus)
Morphological classification of galaxies with deep learning: comparing 3-way and 4-way CNNs
Cavanagh, M. K., Bekki, K. & Groves, B. A., 1 Sept 2021, In: Monthly Notices of the Royal Astronomical Society. 506, 1, p. 659-676 18 p.
Research output: Contribution to journal › Article › peer-review

Open Access

29 Citations (Scopus)

Cite this

@phdthesis{59b2dc70d9b2475bbf51af483b26a70d,

title = "Morphological evolution of galaxies with deep learning",

abstract = "Understanding the changes in galaxy morphology over time can yield important insights into the physics behind galaxy evolution. In this thesis, I adopt a deep learning approach to morphological classification. I develop a new suite of CNNs to rapidly and accurately classify galaxy morphologies. I apply these models to classify tens of thousands of galaxies in several datasets across observational surveys and simulations, analysing the evolution of barred galaxies in EAGLE, the growth of lenticular galaxies over time in COSMOS, and the kinematics of different morphologies in SAMI. I also develop an image segmentation model capable of analysing galaxy bars.",

keywords = "galaxies, morphology, deep learning, morphological classification, machine learning, CNN, galaxy evolution, image classification",

author = "Mitchell Cavanagh",

year = "2023",

doi = "10.26182/ckrn-0t69",

language = "English",

school = "The University of Western Australia",

}

TY - BOOK

T1 - Morphological evolution of galaxies with deep learning

AU - Cavanagh, Mitchell

PY - 2023

Y1 - 2023

N2 - Understanding the changes in galaxy morphology over time can yield important insights into the physics behind galaxy evolution. In this thesis, I adopt a deep learning approach to morphological classification. I develop a new suite of CNNs to rapidly and accurately classify galaxy morphologies. I apply these models to classify tens of thousands of galaxies in several datasets across observational surveys and simulations, analysing the evolution of barred galaxies in EAGLE, the growth of lenticular galaxies over time in COSMOS, and the kinematics of different morphologies in SAMI. I also develop an image segmentation model capable of analysing galaxy bars.

AB - Understanding the changes in galaxy morphology over time can yield important insights into the physics behind galaxy evolution. In this thesis, I adopt a deep learning approach to morphological classification. I develop a new suite of CNNs to rapidly and accurately classify galaxy morphologies. I apply these models to classify tens of thousands of galaxies in several datasets across observational surveys and simulations, analysing the evolution of barred galaxies in EAGLE, the growth of lenticular galaxies over time in COSMOS, and the kinematics of different morphologies in SAMI. I also develop an image segmentation model capable of analysing galaxy bars.

KW - galaxies

KW - morphology

KW - deep learning

KW - morphological classification

KW - machine learning

KW - CNN

KW - galaxy evolution

KW - image classification

U2 - 10.26182/ckrn-0t69

DO - 10.26182/ckrn-0t69

M3 - Doctoral Thesis

ER -

Morphological evolution of galaxies with deep learning

Abstract

Access to Document

Fingerprint

Research output

The redshift evolution of the S0 fraction for z < 1 in COSMOS

The evolution of barred galaxies in the EAGLE simulations

Morphological classification of galaxies with deep learning: comparing 3-way and 4-way CNNs

Cite this