A Bayesian approach to star-galaxy classification

Henrion, Marc; Mortlock, D.; Hand, David J.

doi:10.1111/j.1365-2966.2010.18055.x

Cited by 30 publications

(34 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although star-galaxy separation has traditionally used purely morphometric information to classify stars and galaxies in optical survey data (e.g. Kron 1980;Eisenstein et al 2001;Henrion et al 2011), new ground-based deep-wide surveys, which contain many more unresolved galaxies than stars at faint apparent magnitudes (Fadely et al 2012;Soumagnac et al 2015), represent an emerging challenge. Further work is therefore needed in order to determine whether the algorithm can effectively distinguish faint, unresolved galaxies from stars in very deep images.…”

Section: Star-galaxy Separationmentioning

confidence: 99%

Galaxy morphological classification in deep-wide surveys via unsupervised machine learning

Martin

Kaviraj

Hocking

et al. 2019

Monthly Notices of the Royal Astronomical Society

View full text Add to dashboard Cite

Galaxy morphology is a fundamental quantity, that is essential not only for the full spectrum of galaxy-evolution studies, but also for a plethora of science in observational cosmology (e.g. as a prior for photometric-redshift measurements and as contextual data for transient lightcurve classifications). While a rich literature exists on morphological-classification techniques, the unprecedented data volumes, coupled, in some cases, with the short cadences of forthcoming 'Big-Data' surveys (e.g. from the LSST), present novel challenges for this field. Large data volumes make such datasets intractable for visual inspection (even via massively-distributed platforms like Galaxy Zoo), while short cadences make it difficult to employ techniques like supervised machine-learning, since it may be impractical to repeatedly produce training sets on short timescales. Unsupervised machine learning, which does not require training sets, is ideally suited to the morphological analysis of new and forthcoming surveys. Here, we employ an algorithm that performs clustering of graph representations, in order to group image patches with similar visual properties and objects constructed from those patches, like galaxies. We implement the algorithm on the Hyper-Suprime-Cam Subaru-Strategic-Program Ultra-Deep survey, to autonomously reduce the galaxy population to a small number (160) of 'morphological clusters', populated by galaxies with similar morphologies, which are then benchmarked using visual inspection. The morphological classifications (which we release publicly) exhibit a high level of purity, and reproduce known trends in key galaxy properties as a function of morphological type at z < 1 (e.g. stellar-mass functions, rest-frame colours and the position of galaxies on the star-formation main sequence). Our study demonstrates the power of unsupervised machine learning in performing accurate morphological analysis, which will become indispensable in this new era of deep-wide surveys.

show abstract

Section: Star-galaxy Separationmentioning

confidence: 99%

Galaxy morphological classification in deep-wide surveys via unsupervised machine learning

Martin

Kaviraj

Hocking

et al. 2019

Monthly Notices of the Royal Astronomical Society

View full text Add to dashboard Cite

show abstract

“…Galaxy classification is usually primarily based on detecting source extension by comparing PSF magnitudes with magnitudes based on various profile models. For examples of Galaxy classification, see Vasconcellos et al (2011) who used decision trees for star-galaxy separation in SDSS, or Henrion et al (2011), who used a Bayesian method for star-galaxy separation in SDSS and UKIDSS. Tzalmantza et al (2007Tzalmantza et al ( , 2009 have developed a test library and selection methods for identifying galaxies in the forthcoming Gaia mission.…”

Section: Svm For Pandiscmentioning

confidence: 99%

THE PHOTOMETRIC CLASSIFICATION SERVER FOR Pan-STARRS1

Saglia¹,

Tonry²,

Bender³

et al. 2012

ApJ

View full text Add to dashboard Cite

The Pan-STARRS1 survey is obtaining multi-epoch imaging in five bands (g P1 r P1 i P1 z P1 y P1 ) over the entire sky north of declination −30 deg. We describe here the implementation of the Photometric Classification Server (PCS) for Pan-STARRS1. PCS will allow the automatic classification of objects into star/galaxy/quasar classes based on colors and the measurement of photometric redshifts for extragalactic objects, and will constrain stellar parameters for stellar objects, working at the catalog level. We present tests of the system based on high signal-to-noise photometry derived from the Medium-Deep Fields of Pan-STARRS1, using available spectroscopic surveys as training and/or verification sets. We show that the Pan-STARRS1 photometry delivers classifications and photometric redshifts as good as the Sloan Digital Sky Survey (SDSS) photometry to the same magnitude limits. In particular, our preliminary results, based on this relatively limited data set down to the SDSS spectroscopic limits, and therefore potentially improvable, show that stars are correctly classified as such in 85% of cases, galaxies in 97%, and QSOs in 84%. False positives are less than 1% for galaxies, ≈19% for stars, and ≈28% for QSOs. Moreover, photometric redshifts for 1000 luminous red galaxies up to redshift 0.5 are determined to 2.4% precision (defined as 1.48 × Median|z phot − z spec |/(1 + z)) with just 0.4% catastrophic outliers and small (−0.5%) residual bias. For bluer galaxies up to the same redshift, the residual bias (on average −0.5%) trend, percentage of catastrophic failures (1.2%), and precision (4.2%) are higher, but still interestingly small for many science applications. Good photometric redshifts (to 5%) can be obtained for at most 60% of the QSOs of the sample. PCS will create a value-added catalog with classifications and photometric redshifts for eventually many millions of sources.

show abstract

“…With typical seeing and noise conditions in these images, small, faint galaxies become indistinguishable from stars. A wide range of techniques has been developed to resolve this problem (Henrion et al 2011;Fadely et al 2012;Soumagnac et al 2013). Standard star-galaxy classifiers use morphological information of the stars, more advanced ones incorporate also the color information (Pollo et al 2010).…”

Section: Star-galaxy Classificationmentioning

confidence: 99%

Modeling the Transfer Function for the Dark Energy Survey

Chang¹,

Busha

Wechsler

et al. 2015

ApJ

View full text Add to dashboard Cite

We present a forward-modeling simulation framework designed to model the data products from the Dark Energy Survey (DES). This forward-model process can be thought of as a transfer function-a mapping from cosmological/ astronomical signals to the final data products used by the scientists. Using output from the cosmological simulations (the Blind Cosmology Challenge), we generate simulated images (the Ultra Fast Image Simulator) and catalogs representative of the DES data. In this work we demonstrate the framework by simulating the 244 deg 2 coadd images and catalogs in five bands for the DES Science Verification data. The simulation output is compared with the corresponding data to show that major characteristics of the images and catalogs can be captured. We also point out several directions of future improvements. Two practical examples-star-galaxy classification and proximity effects on object detection-are then used to illustrate how one can use the simulations to address systematics issues in data analysis. With clear understanding of the simplifications in our model, we show that one can use the simulations side-by-side with data products to interpret the measurements. This forward modeling approach is generally applicable for other upcoming and future surveys. It provides a powerful tool for systematics studies that is sufficiently realistic and highly controllable.

show abstract

A Bayesian approach to star-galaxy classification

Cited by 30 publications

References 37 publications

Galaxy morphological classification in deep-wide surveys via unsupervised machine learning

Galaxy morphological classification in deep-wide surveys via unsupervised machine learning

THE PHOTOMETRIC CLASSIFICATION SERVER FOR Pan-STARRS1

Modeling the Transfer Function for the Dark Energy Survey

Contact Info

Product

Resources

About