A Comparison of Deep Learning Architectures for Optical Galaxy Morphology Classification (2021-11-08T00:00:00.000000Z)

TL;DR

A comparison of deep learning architectures to determine which is best suited for optical galaxy morphology classification is offered, and DenseNet121 was found to produce the best results, in terms of accuracy, with a reasonable training time.

Abstract

The classification of galaxy morphology plays a crucial role in understanding galaxy formation and evolution. Traditionally, this process is done manually. The emergence of deep learning techniques has given room for the automation of this process. As such, this paper offers a comparison of deep learning architectures to determine which is best suited for optical galaxy morphology classification. Adapting the model training method proposed by Walmsley et al in 2021, the Zoobot Python library is used to train models to predict Galaxy Zoo DECaLS decision tree responses, made by volunteers, using EfficientNet B0, DenseNet121 and ResNet50 as core model architectures. The predicted results are then used to generate accuracy metrics per decision tree question to determine architecture performance. DenseNet121 was found to produce the best results, in terms of accuracy, with a reasonable training time. In future, further testing with more deep learning architectures could prove beneficial.

Authors

Ezra Fielding

2 papers

Clement N. Nyirenda

2 papers

Mattia Vaccari

2 papers

A Comparison of Deep Learning Architectures for Optical Galaxy Morphology Classification

TL;DR

Abstract

Authors

References28 items

Square Kilometre Array

A Comparative Evaluation of Spatio Temporal Deep Learning Techniques for Crime Prediction

An Efficient DenseNet-Based Deep Learning Model for Malware Detection

Galaxy Zoo DECaLS: Detailed Visual Morphology Measurements from Volunteers and Deep Learning for 314, 000 Galaxies

CNN architecture comparison for radio galaxy classification

Skin Cancer Classification using ResNet

3D DenseNet Ensemble in 4-Way Classification of Alzheimer's Disease

Automated medical diagnosis of COVID-19 through EfficientNet convolutional neural network

Channel-Attention-Based DenseNet Network for Remote Sensing Image Scene Classification

Cataloguing the radio-sky with unsupervised machine learning: a new approach for the SKA era

Automated fruit recognition using EfficientNet and MixNet

Surveying the reach and maturity of machine learning and artificial intelligence in astronomy

Deblending and classifying astronomical sources with Mask R-CNN deep learning

Twelve years of Galaxy Zoo

EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks

Galaxy Zoo: Probabilistic Morphology through Bayesian CNNs and Active Learning

A Machine Learning Based Morphological Classification of 14,245 Radio AGNs Selected from the Best–Heckman Sample

The FIRST Classifier: compact and extended radio galaxy classification using deep Convolutional Neural Networks

Radio Galaxy Zoo: compact and extended radio source classification with deep learning

Deep Learning for Real-Time Crime Forecasting and Its Ternarization

Improving galaxy morphologies for SDSS with Deep Learning

Densely Connected Convolutional Networks

Deep Residual Learning for Image Recognition

Radio Galaxy Zoo : host galaxies and radio morphologies derived from visual inspection

Rotation-invariant convolutional neural networks for galaxy morphology prediction

Galaxy Zoo: morphologies derived from visual inspection of galaxies from the Sloan Digital Sky Survey

“Zoobot,”

Multi-Label Classification of Fundus Images With EfficientNet

Field of Study

Journal Information

Name

Page