DeepShadows: Finding low-surface-brightness galaxies in survey images
Abstract: DeepShadows uses a convolutional neural networks (CNNs) to separate low-surface-brightness galaxies (LSBGs) from artifacts (such as Galactic cirrus and star-forming regions) in survey images. The model is trained and tested on labeled LSBGs and artifacts from the Dark Energy Survey and demonstrates that CNNs offer a promising path in the quest to study the low-surface-brightness universe.
Credit: Tanoglidis, Dimitrios; Ćiprijanović, Aleksandra; Drlica-Wagner, Alex
Site: https://github.com/dtanoglidis/DeepShadows
https://ui.adsabs.harvard.edu/abs/2020arXiv201112437T
Bibcode: 2020ascl.soft11026T
ID: ascl:2011.026