Panoptic Segmentation of Galactic Structures in LSB Images - Archive ouverte HAL
Communication Dans Un Congrès Année : 2023

Panoptic Segmentation of Galactic Structures in LSB Images

Résumé

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.
Fichier principal
Vignette du fichier
Multi_class_Segmentation_of_Galactic_Structures.pdf (5.68 Mo) Télécharger le fichier
Licence
Copyright (Tous droits réservés)

Dates et versions

hal-04129549 , version 1 (27-06-2023)

Licence

Copyright (Tous droits réservés)

Identifiants

  • HAL Id : hal-04129549 , version 1

Citer

Felix Richards, Adeline Paiement, Xianghua Xie, Elisabeth Sola, Pierre-Alain Duc. Panoptic Segmentation of Galactic Structures in LSB Images. 18th International Conference on Machine Vision Applications, Jul 2023, Hamamatsu, Shizuoka, Japan. ⟨hal-04129549⟩
28 Consultations
123 Téléchargements

Partager

More