Evaluation of Galaxy Morphology Classification with Machine Learning and Deep Learning

DOI

10.2991/978-94-6239-648-7_63How to use a DOI?

Keywords

Galaxy Morphology Classification; Deep Learning in Astronomy; Convolutional Neural Networks (CNNs); Galaxy10 DECaLS Dataset; Machine Learning for Sky Surveys

Abstract

Through the development of technologies, astronomical imaging surveys have increased significantly, so that more galaxy images are taken than ever, which makes the traditional manual galaxy classification infeasible. For this reason, adopting automated machine learning techniques is important in replacing the traditional way of galaxy classification that requires cooperation with large-scale sky surveys. This research examines the performance of traditional and deep learning models through classifying galaxies from ten morphological features. The study used the Galaxy10 DECaLS dataset, which contained 17,736 colored galaxy images with labels. The dataset was first normalized and separated into training, validation, and testing subsets in proportions of 70, 15, and 15 percent. Then, the subsets of data were fed to the models. Traditional models like Support Vector Machine (SVM) with an RBF kernel and Random Forest with 500 estimators were applied to PCA-compressed features, and neural architectures like Multilayer Perceptron (MLP) and Convolutional Neural Network (CNN), were trained directly on images. The result indicates that the deep learning models substantially outperform the classical methods. CNN achieved the highest accuracies and best generalization to unseen data, which confirms the advantage of convolutional feature extraction over hand-engineered representations. However, deep learning models require large computing power and a dataset to reach an ideal high accuracy. Those findings show that the deep learning models show scalable, accurate morphological classification, which offers significant potential for future large-scale astronomical surveys.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Yuhan You
PY  - 2026
DA  - 2026/04/24
TI  - Evaluation of Galaxy Morphology Classification with Machine Learning and Deep Learning
BT  - Proceedings of the International Workshop on Advances in Deep Learning for Image Analysis and Computer Vision (IWADIC 2025)
PB  - Atlantis Press
SP  - 571
EP  - 582
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6239-648-7_63
DO  - 10.2991/978-94-6239-648-7_63
ID  - You2026
ER  -