Galaxy morphology prediction using capsule networks

TL;DR

This paper explores the use of Capsule Networks for galaxy morphology classification, leveraging their rotational invariance to improve accuracy and reduce astronomers' workload in large sky surveys.

Contribution

It demonstrates the effectiveness of Capsule Networks in galaxy morphology classification through two evaluation scenarios, including regression and classification with image reconstruction.

Findings

Capsule Networks achieved promising classification results.

The approach reduces manual effort in galaxy morphology analysis.

Rotational invariance of Capsule Networks benefits astronomical image classification.

Abstract

Understanding morphological types of galaxies is a key parameter for studying their formation and evolution. Neural networks that have been used previously for galaxy morphology classification have some disadvantages, such as not being invariant under rotation. In this work, we studied the performance of Capsule Network, a recently introduced neural network architecture that is rotationally invariant and spatially aware, on the task of galaxy morphology classification. We designed two evaluation scenarios based on the answers from the question tree in the Galaxy Zoo project. In the first scenario, we used Capsule Network for regression and predicted probabilities for all of the questions. In the second scenario, we chose the answer to the first morphology question that had the highest user agreement as the class of the object and trained a Capsule Network classifier, where we also reconstructed galaxy images. We achieved promising results in both of these scenarios. Automated approaches such as the one introduced here will greatly decrease the workload of astronomers and will play a critical role in the upcoming large sky surveys.