Deep Transfer Learning for Star Cluster Classification: I. Application to the PHANGS-HST Survey

TL;DR

This study demonstrates that deep transfer learning models can effectively classify star clusters in galaxy images, achieving accuracy comparable to human experts, and lays the groundwork for automated, large-scale star cluster classification.

Contribution

The paper introduces a transfer learning approach for star cluster classification, showing its effectiveness and robustness across different neural network architectures and training datasets.

Findings

Classification accuracy of ~70% for certain classes

Performance comparable to human expert classifications

Robustness across different neural network architectures

Abstract

We present the results of a proof-of-concept experiment which demonstrates that deep learning can successfully be used for production-scale classification of compact star clusters detected in HST UV-optical imaging of nearby spiral galaxies (D < 20 Mpc) in the PHANGS-HST survey. Given the relatively small nature of existing, human-labelled star cluster samples, we transfer the knowledge of state-of-the-art neural network models for real-object recognition to classify star clusters candidates into four morphological classes. We perform a series of experiments to determine the dependence of classification performance on: neural network architecture (ResNet18 and VGG19-BN); training data sets curated by either a single expert or three astronomers; and the size of the images used for training. We find that the overall classification accuracies are not significantly affected by these choices. The networks are used to classify star cluster candidates in the PHANGS-HST galaxy NGC 1559, which was not included in the training samples. The resulting prediction accuracies are 70%, 40%, 40-50%, 50-70% for class 1, 2, 3 star clusters, and class 4 non-clusters respectively. This performance is competitive with consistency achieved in previously published human and automated quantitative classification of star cluster candidate samples (70-80%, 40-50%, 40-50%, and 60-70%). The methods introduced herein lay the foundations to automate classification for star clusters at scale, and exhibit the need to prepare a standardized dataset of human-labelled star cluster classifications, agreed upon by a full range of experts in the field, to further improve the performance of the networks introduced in this study.