J/MNRAS/483/529 Star-galaxy multi narrow-band classification (Cabayol+, 2019)
The PAU survey: star-galaxy classification with multi narrow-band data.
Cabayol L., Sevilla-Noarbe I., Fernandez E., Carretero J., Eriksen M.,
Serrano S., Alarcon A., Amara A., Casas R., Castander F.J., de Vicente J.,
Folger M., Garcia-Bellido J., Gaztanaga E., Hoekstra H., Miquel R.,
Padilla C., Sanchez E., Stothert L., Tallada P., Tortorelli L.
<Mon. Not. R. Astron. Soc., 483, 529-539 (2019)>
=2019MNRAS.483..529C 2019MNRAS.483..529C (SIMBAD/NED BibCode)
ADC_Keywords: Photometry, narrow-band ; Photometry, classification ; Morphology
Keywords: methods: data analysis, techniques: photometric
Abstract:
Classification of stars and galaxies is a well-known astronomical
problem that has been treated using different approaches, most of them
relying on morphological information. In this paper, we tackle this
issue using the low-resolution spectra from narrow-band photometry,
provided by the Physics of the Accelerating Universe survey. We find
that, with the photometric fluxes from the 40 narrow-band filters and
without including morphological information, it is possible to
separate stars and galaxies to very high precision, 98.4 per cent
purity with a completeness of 98.8 per cent for objects brighter than
I=22.5. This precision is obtained with a convolutional neural network
as a classification algorithm, applied to the objects' spectra. We
have also applied the method to the ALHAMBRA photometric survey and we
provide an updated classification for its Gold sample.
Description:
In this work, we would like to assess the performance of a machine
learning classifier over two narrow-band data sets, PAUS early data
and the ALHAMBRA Gold catalogue (Molino et al. 2014MNRAS.441.2891M 2014MNRAS.441.2891M,
Cat. J/MNRAS/441/2891), in the latter case comparing with the standard
classification provided by that survey. We will work on the COSMOS
field comparing against the COSMOS space-based imaging catalogue
(Leauthaud et al. 2007ApJS..172..219L 2007ApJS..172..219L), which provides a
morphology-based classification for the objects to train and test our
methods on. It contains 1.2x106 objects to a limiting magnitude of
F814W=26.5 from images observed with the Hubble Space Telescope (HST)
using the Advanced Camera for Surveys (ACS), therefore its image
quality (very deep and unaffected by the atmosphere) can be used as a
'truth' reference. As a part of this work, we provide an additional
column for the ALHAMBRA Gold data set (Molino et al.
2014MNRAS.441.2891M 2014MNRAS.441.2891M, Cat. J/MNRAS/441/2891) for which we have computed
the stellarity value developed in this paper. The fields covered are
from ALHAMBRA-2 to ALHAMBRA-8, in correspondence to DEEP-2, SDSS,
COSMOS, HDF-N, GROTH, ELAIS-N1, and SDSS, respectively. The catalogue
with this classification is available at http://cosmohub.pic.es.
In Table A1 we present the value-added catalogue columns that are
being provided (most inherited from the original Gold catalogue, for
reference).
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
tablea1.dat 98 421874 The ALHAMBRA catalogue extension with CNN
classification
--------------------------------------------------------------------------------
See also:
J/MNRAS/441/2891 : ALHAMBRA Survey (Molino+, 2014)
Byte-by-byte Description of file: tablea1.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 11 I11 --- ID ALHAMBRA's unique object identifier
13- 20 F8.4 deg RAdeg Right ascension (J2000)
22- 28 F7.4 deg DEdeg Declination (J2000)
30- 33 F4.2 --- Starflag ALHAMBRA's Statistical STAR/GALAXY
Discriminator (0:Pure-Galaxy, 0.5:Unknown,
1:Pure-Star)
35- 52 F18.15 mag F814Wmag Isophotal magnitude (AB)
54- 75 F22.18 mag e_F814Wmag [] Error on F814Wmag
77- 98 E22.19 --- cnn CNN star/galaxy discriminator probability
(0:Pure-Galaxy,1:Pure-Star) (1)
--------------------------------------------------------------------------------
Note (1): We only provide a classification for those objects with all bands
measured (20 optical, three NIR, and F814W). For those without, the
class is set to a 'sentinel' value of -1.
--------------------------------------------------------------------------------
History:
From electronic version of the journal
(End) Ana Fiallos [CDS] 06-Jul-2022