Deep Learning Segmentation of Spiral Arms and Bars

TL;DR

This paper introduces the first deep learning model for segmenting galactic spiral arms and bars, outperforming existing automated methods and expert labels, enabling large-scale astrophysical research.

Contribution

A novel deep learning approach for galactic structure segmentation, validated by expert assessments and crowdsourcing, advancing automated analysis in astronomy.

Findings

Predicted spiral arm masks preferred over existing methods (99%)

Expert-rated masks are mostly good to perfect (89%)

Bar length measurements agree with crowdsourcing data

Abstract

We present the first deep learning model for segmenting galactic spiral arms and bars. In a blinded assessment by expert astronomers, our predicted spiral arm masks are preferred over both current automated methods (99% of evaluations) and our original volunteer labels (79% of evaluations). Experts rated our spiral arm masks as `mostly good' to `perfect' in 89% of evaluations. Bar lengths trivially derived from our predicted bar masks are in excellent agreement with a dedicated crowdsourcing project. The pixelwise precision of our masks, previously impossible at scale, will underpin new research into how spiral arms and bars evolve.