Panoptic Segmentation of Galactic Structures in LSB Images

TL;DR

This work addresses automated localization and segmentation of galactic structures in low surface brightness images, where cirrus contamination and artefacts hinder traditional detection. It proposes a panoptic segmentation framework that combines Mask R-CNN with a cirrus-specific semantic network and augments it with an adaptive intensity scaling layer and a human-in-the-loop training scheme. The approach yields notable gains over separate baselines, particularly for diffuse and ghosted halos, and demonstrates strong improvement in cirrus segmentation due to shared backbone features. The HITL training protocol substantially enhances segmentation accuracy when data are limited, highlighting its practical value for astronomical image analysis and cataloguing in large surveys.

Abstract

We explore the use of deep learning to localise galactic structures in low surface brightness (LSB) images. LSB imaging reveals many interesting structures, though these are frequently confused with galactic dust contamination, due to a strong local visual similarity. We propose a novel unified approach to multi-class segmentation of galactic structures and of extended amorphous image contaminants. Our panoptic segmentation model combines Mask R-CNN with a contaminant specialised network and utilises an adaptive preprocessing layer to better capture the subtle features of LSB images. Further, a human-in-the-loop training scheme is employed to augment ground truth labels. These different approaches are evaluated in turn, and together greatly improve the detection of both galactic structures and contaminants in LSB images.

Figures (1)

• Figure 1: Proposed panoptic segmentation model.