An official website of the United States government
e apply a new deep learning technique to detect, classify, and deblend sources in multi-band astronomical images. We train and evaluate the performance of an artificial neural network built on the Mask R-CNN image processing framework, a general code for efficient object detection, classification, and instance segmentation. After evaluating the performance of our network against simulated ground truth images for star and galaxy classes, we find a precision of 92% at 80% recall for stars and a precision of 98% at 80% recall for galaxies in a typical field with ∼30 galaxies/arcmin2. We investigate the deblending capability of our code, and find that clean deblends are handled robustly during object masking, even for significantly blended sources. This technique, or extensions using similar network architectures, may be applied to current and future deep imaging surveys such as LSST and WFIRST. Our code, Astro R-CNN, is publicly available at https://github.com/burke86/astro_rcnn
more »
« less
Detection, instance segmentation, and classification for astronomical surveys with deep learning ( deepdisc ): detectron2 implementation and demonstration with Hyper Suprime-Cam data
ABSTRACT The next generation of wide-field deep astronomical surveys will deliver unprecedented amounts of images through the 2020s and beyond. As both the sensitivity and depth of observations increase, more blended sources will be detected. This reality can lead to measurement biases that contaminate key astronomical inferences. We implement new deep learning models available through Facebook AI Research’s detectron2 repository to perform the simultaneous tasks of object identification, deblending, and classification on large multiband co-adds from the Hyper Suprime-Cam (HSC). We use existing detection/deblending codes and classification methods to train a suite of deep neural networks, including state-of-the-art transformers. Once trained, we find that transformers outperform traditional convolutional neural networks and are more robust to different contrast scalings. Transformers are able to detect and deblend objects closely matching the ground truth, achieving a median bounding box Intersection over Union of 0.99. Using high-quality class labels from the Hubble Space Telescope, we find that when classifying objects as either stars or galaxies, the best-performing networks can classify galaxies with near 100 per cent completeness and purity across the whole test sample and classify stars above 60 per cent completeness and 80 per cent purity out to HSC i-band magnitudes of 25 mag. This framework can be extended to other upcoming deep surveys such as the Legacy Survey of Space and Time and those with the Roman Space Telescope to enable fast source detection and measurement. Our code, deepdisc, is publicly available at https://github.com/grantmerz/deepdisc.
more »
« less
- PAR ID:
- 10465660
- Publisher / Repository:
- Oxford University Press
- Date Published:
- Journal Name:
- Monthly Notices of the Royal Astronomical Society
- Volume:
- 526
- Issue:
- 1
- ISSN:
- 0035-8711
- Format(s):
- Medium: X Size: p. 1122-1137
- Size(s):
- p. 1122-1137
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
ABSTRACT We present a mock image catalogue of ∼100 000 MUV ≃ −22.5 to −19.6 mag galaxies at z = 7–12 from the bluetides cosmological simulation. We create mock images of each galaxy with the James Webb Space Telescope (JWST), Hubble, Roman, and Euclid Space Telescopes, as well as Subaru, and VISTA, with a range of near- and mid-infrared filters. We perform photometry on the mock images to estimate the success of these instruments for detecting high-z galaxies. We predict that JWST will have unprecedented power in detecting high-z galaxies, with a 95 per cent completeness limit at least 2.5 mag fainter than VISTA and Subaru, 1.1 mag fainter than Hubble, and 0.9 mag fainter than Roman, for the same wavelength and exposure time. Focusing on JWST, we consider a range of exposure times and filters, and find that the NIRCam F356W and F277W filters will detect the faintest galaxies, with 95 per cent completeness at m ≃ 27.4 mag in 10-ks exposures. We also predict the number of high-z galaxies that will be discovered by upcoming JWST imaging surveys. We predict that the COSMOS-Web survey will detect ∼1000 M1500 Å < −20.1 mag galaxies at 6.5 < z < 7.5, by virtue of its large survey area. JADES-Medium will detect almost $$100{{\ \rm per\ cent}}$$ of M1500 Å ≲ −20 mag galaxies at z < 8.5 due to its significant depth, however, with its smaller survey area it will detect only ∼100 of these galaxies at 6.5 < z < 7.5. Cosmic variance results in a large range in the number of predicted galaxies each survey will detect, which is more evident in smaller surveys such as CEERS and the PEARLS NEP and GOODS-S fields.more » « less
-
ABSTRACT We present the results of a proof-of-concept experiment that demonstrates that deep learning can successfully be used for production-scale classification of compact star clusters detected in Hubble Space Telescope(HST) ultraviolet-optical imaging of nearby spiral galaxies ($$D\lesssim 20\, \textrm{Mpc}$$) in the Physics at High Angular Resolution in Nearby GalaxieS (PHANGS)–HST survey. Given the relatively small nature of existing, human-labelled star cluster samples, we transfer the knowledge of state-of-the-art neural network models for real-object recognition to classify star clusters candidates into four morphological classes. We perform a series of experiments to determine the dependence of classification performance on neural network architecture (ResNet18 and VGG19-BN), training data sets curated by either a single expert or three astronomers, and the size of the images used for training. We find that the overall classification accuracies are not significantly affected by these choices. The networks are used to classify star cluster candidates in the PHANGS–HST galaxy NGC 1559, which was not included in the training samples. The resulting prediction accuracies are 70 per cent, 40 per cent, 40–50 per cent, and 50–70 per cent for class 1, 2, 3 star clusters, and class 4 non-clusters, respectively. This performance is competitive with consistency achieved in previously published human and automated quantitative classification of star cluster candidate samples (70–80 per cent, 40–50 per cent, 40–50 per cent, and 60–70 per cent). The methods introduced herein lay the foundations to automate classification for star clusters at scale, and exhibit the need to prepare a standardized data set of human-labelled star cluster classifications, agreed upon by a full range of experts in the field, to further improve the performance of the networks introduced in this study.more » « less
-
Astronomical source deblending is the process of separating the contribution of individual stars or galaxies (sources) to an image comprised of multiple, possibly overlapping sources. Astronomical sources display a wide range of sizes and brightnesses and may show substantial overlap in images. Astronomical imaging data can further challenge off-the-shelf computer vision algorithms owing to its high dynamic range, low signal-to-noise ratio, and unconventional image format. These challenges make source deblending an open area of astronomical research, and in this work, we introduce a new approach called Partial-Attribution Instance Segmentation that enables source detection and deblending in a manner tractable for deep learning models. We provide a novel neural network implementation as a demonstration of the method.more » « less
-
ABSTRACT Very metal-poor stars ($$\rm [Fe/H] \lt -2$$) in the Milky Way are fossil records of early chemical evolution and the assembly and structure of the Galaxy. However, they are rare and hard to find. Gaia DR3 has provided over 200 million low-resolution (R ≈ 50) XP spectra, which provides an opportunity to greatly increase the number of candidate metal-poor stars. In this work, we utilize the XGBoost classification algorithm to identify ∼200 000 very metal-poor star candidates. Compared to past work, we increase the candidate metal-poor sample by about an order of magnitude, with comparable or better purity than past studies. First, we develop three classifiers for bright stars (BP < 16). They are Classifier-T (for Turn-off stars), Classifier-GC (for Giant stars with high completeness), and Classifier-GP (for Giant stars with high purity) with expected purity of 52 per cent/45 per cent/76 per cent and completeness of 32 per cent/93 per cent/66 per cent, respectively. These three classifiers obtained a total of 11 000/111 000/44 000 bright metal-poor candidates. We apply model-T and model-GP on faint stars (BP > 16) and obtain 38 000/41 000 additional metal-poor candidates with purity 29 per cent/52 per cent, respectively. We make our metal-poor star catalogues publicly available, for further exploration of the metal-poor Milky Way.more » « less
