skip to main content

Title: Detecting Low Surface Brightness Galaxies with Mask R-CNN
Low surface brightness galaxies (LSBGs), galaxies that are fainter than the dark night sky, are famously difficult to detect. Nonetheless, studies of these galaxies are essential to improve our understanding of the formation and evolution of low-mass galaxies. In this work, we train a deep learning model using the Mask R-CNN framework on a set of simulated LSBGs inserted into images from the Dark Energy Survey (DES) Data Release 2 (DR2). This deep learning model is combined with several conventional image pre-processing steps to develop a pipeline for the detection of LSBGs. We apply this pipeline to the full DES DR2 coadd image dataset, and preliminary results show the detection of 22 large, high-quality LSBG candidates that went undetected by conventional algorithms. Furthermore, we find that the performance of our algorithm is greatly improved by including examples of false positives as an additional class during training.
; ; ; ; ;
Award ID(s):
Publication Date:
Journal Name:
Workshop at the 35th Conference on Neural Information Processing Systems (NeurIPS)
Page Range or eLocation-ID:
Sponsoring Org:
National Science Foundation
More Like this

    In this work, we explore the possibility of applying machine learning methods designed for 1D problems to the task of galaxy image classification. The algorithms used for image classification typically rely on multiple costly steps, such as the point spread function deconvolution and the training and application of complex Convolutional Neural Networks of thousands or even millions of parameters. In our approach, we extract features from the galaxy images by analysing the elliptical isophotes in their light distribution and collect the information in a sequence. The sequences obtained with this method present definite features allowing a direct distinction between galaxy types. Then, we train and classify the sequences with machine learning algorithms, designed through the platform Modulos AutoML. As a demonstration of this method, we use the second public release of the Dark Energy Survey (DES DR2). We show that we are able to successfully distinguish between early-type and late-type galaxies, for images with signal-to-noise ratio greater than 300. This yields an accuracy of $86{{\ \rm per\ cent}}$ for the early-type galaxies and $93{{\ \rm per\ cent}}$ for the late-type galaxies, which is on par with most contemporary automated image classification approaches. The data dimensionality reduction of our novelmore »method implies a significant lowering in computational cost of classification. In the perspective of future data sets obtained with e.g. Euclid and the Vera Rubin Observatory, this work represents a path towards using a well-tested and widely used platform from industry in efficiently tackling galaxy classification problems at the peta-byte scale.

    « less
  2. Abstract We present morphological classifications of ∼27 million galaxies from the Dark Energy Survey (DES) Data Release 1 (DR1) using a supervised deep learning algorithm. The classification scheme separates: (a) early-type galaxies (ETGs) from late-types (LTGs), and (b) face-on galaxies from edge-on. Our Convolutional Neural Networks (CNNs) are trained on a small subset of DES objects with previously known classifications. These typically have mr ≲ 17.7mag; we model fainter objects to mr < 21.5 mag by simulating what the brighter objects with well determined classifications would look like if they were at higher redshifts. The CNNs reach 97% accuracy to mr < 21.5 on their training sets, suggesting that they are able to recover features more accurately than the human eye. We then used the trained CNNs to classify the vast majority of the other DES images. The final catalog comprises five independent CNN predictions for each classification scheme, helping to determine if the CNN predictions are robust or not. We obtain secure classifications for ∼ 87% and 73% of the catalog for the ETG vs. LTG and edge-on vs. face-on models, respectively. Combining the two classifications (a) and (b) helps to increase the purity of the ETG sample andmore »to identify edge-on lenticular galaxies (as ETGs with high ellipticity). Where a comparison is possible, our classifications correlate very well with Sérsic index (n), ellipticity (ε) and spectral type, even for the fainter galaxies. This is the largest multi-band catalog of automated galaxy morphologies to date.« less

    We introduce the southern stellar stream spectroscopy survey (S5), an on-going program to map the kinematics and chemistry of stellar streams in the southern hemisphere. The initial focus of S5 has been spectroscopic observations of recently identified streams within the footprint of the dark energy survey (DES), with the eventual goal of surveying streams across the entire southern sky. Stellar streams are composed of material that has been tidally striped from dwarf galaxies and globular clusters and hence are excellent dynamical probes of the gravitational potential of the Milky Way, as well as providing a detailed snapshot of its accretion history. Observing with the 3.9 m Anglo-Australian Telescope’s 2-degree-Field fibre positioner and AAOmega spectrograph, and combining the precise photometry of DES DR1 with the superb proper motions from Gaia DR2, allows us to conduct an efficient spectroscopic survey to map these stellar streams. So far S5 has mapped nine DES streams and three streams outside of DES; the former are the first spectroscopic observations of these recently discovered streams. In addition to the stream survey, we use spare fibres to undertake a Milky Way halo survey and a low-redshift galaxy survey. This paper presents an overview of the S5 program,more »describing the scientific motivation for the survey, target selection, observation strategy, data reduction, and survey validation. Finally, we describe early science results on stellar streams and Milky Way halo stars drawn from the survey. Updates on S5, including future public data releases, can be found at

    « less
  4. Wide-field astronomical surveys are often affected by the presence of undesirable reflections (often known as “ghosting artifacts” or “ghosts”) and scattered-light artifacts. The identification and mitigation of these artifacts is important for rigorous astronomical analyses of faint and low-surface-brightness systems. In this work, we use images from the Dark Energy Survey (DES) to train, validate, and test a deep neural network (Mask R-CNN) to detect and localize ghosts and scatteredlight artifacts. We find that the ability of the Mask R-CNN model to identify affected regions is superior to that of conventional algorithms that model the physical processes that lead to such artifacts, thus providing a powerful technique for the automated detection of ghosting and scattered-light artifacts in current and near-future surveys.
  5. ABSTRACT The Cold Spot is a puzzling large-scale feature in the Cosmic Microwave Background temperature maps and its origin has been subject to active debate. As an important foreground structure at low redshift, the Eridanus supervoid was recently detected, but it was subsequently determined that, assuming the standard ΛCDM model, only about 10–20 per cent of the observed temperature depression can be accounted for via its Integrated Sachs–Wolfe imprint. However, R ≳ 100 h−1Mpc supervoids elsewhere in the sky have shown ISW imprints AISW ≈ 5.2 ± 1.6 times stronger than expected from ΛCDM (AISW = 1), which warrants further inspection. Using the Year-3 redMaGiC catalogue of luminous red galaxies from the Dark Energy Survey, here we confirm the detection of the Eridanus supervoid as a significant underdensity in the Cold Spot’s direction at z < 0.2. We also show, with S/N ≳ 5 significance, that the Eridanus supervoid appears as the most prominent large-scale underdensity in the dark matter mass maps that we reconstructed from DES Year-3 gravitational lensing data. While we report no significant anomalies, an interesting aspect is that the amplitude of the lensing signal from the Eridanus supervoid at the Cold Spot centre is about 30 per cent lower thanmore »expected from similar peaks found in N-body simulations based on the standard ΛCDM model with parameters Ωm = 0.279 and σ8 = 0.82. Overall, our results confirm the causal relation between these individually rare structures in the cosmic web and in the CMB, motivating more detailed future surveys in the Cold Spot region.« less