Machine learning technique for morphological classification of galaxies from the SDSS. III. Image-based inference of detailed features

Abstract

This paper follows series of our works on the applicability of various machine learning methods to the morphological galaxy classification (Vavilova et al., 2021, 2022). We exploited the sample of 315776 SDSS DR9 galaxies with absolute stellar magnitudes of -24m<Mr<-19.4m at 0.003<z<0.1 as a target data set for the CNN classifier based on the DenseNet-201. Because it is tightly overlapped with the Galaxy Zoo 2 (GZ2) sample, we use these annotated data as the training data set to classify galaxies into 34 detailed features. In the presence of a pronounced difference of visual parameters between galaxies from the GZ2 training data set and galaxies without known morphological parameters, we applied novel procedures, which allowed us for the first time to get rid of this difference for smaller and fainter SDSS galaxies. We describe in detail the adversarial validation technique as well as how we managed the optimal train-test split of galaxies from the training data set. We have also found optimal galaxy image transformations to increase the classifier generalization ability. It can be considered as another way to improve the human bias for those galaxy images that had a poor vote classification in the GZ project. Such an approach, likely auto-immunization, when the CNN classifier trained on very good images is able to retrain bad images from the same homogeneous sample, can be considered co-planar to other methods of combating the human bias. The accuracy of CNN classifier is in the range of 83.3-99.4 percent depending on 32 features. As a result, for the first time, we assigned the detailed morphological classification for more than 140K low-redshift galaxies, especially at the fainter end. We accentuate on the typical problem points of galaxy CNN image classification from the astronomical point of view. The catalogs will be available through the VizieR.

arXiv comment42 pages, 4 tables, 13 figures, 122 references

References in corpus (86)

• The SIMBAD astronomical database
• Sloan Digital Sky Survey IV: Mapping the Milky Way, Nearby Galaxies, and the Distant Universe
• The Ninth Data Release of the Sloan Digital Sky Survey: First Spectroscopic Data from the SDSS-III Baryon Oscillation Spectroscopic Survey
• Rotation-invariant convolutional neural networks for galaxy morphology prediction
• The Green Valley is a Red Herring: Galaxy Zoo reveals two evolutionary pathways towards quenching of star formation in early- and late-type galaxies
• Galaxy Zoo 2: detailed morphological classifications for 304,122 galaxies from the Sloan Digital Sky Survey
• The Luminosity Function of Galaxies in SDSS Commissioning Data
• Preparing Red-Green-Blue (RGB) Images from CCD Data
• Improving galaxy morphologies for SDSS with Deep Learning
• Galaxy Number Counts from the Sloan Digital Sky Survey Commissioning Data
• Generative Adversarial Networks recover features in astrophysical images of galaxies beyond the deconvolution limit
• Galaxy Zoo: the fraction of merging galaxies in the SDSS and their morphologies
• ANNz2 - photometric redshift and probability distribution function estimation using machine learning
• CMU DeepLens: Deep Learning For Automatic Image-based Galaxy-Galaxy Strong Lens Finding
• Photometric redshifts from SDSS images using a Convolutional Neural Network
• Galaxy Zoo: Probabilistic Morphology through Bayesian CNNs and Active Learning
• Redshift distributions of galaxies in the DES Science Verification shear catalogue and implications for weak lensing
• Classifying Radio Galaxies with Convolutional Neural Network
• Galaxy Zoo: comparing the demographics of spiral arm number and a new method for correcting redshift bias
• Galaxy Morphology Classification with Deep Convolutional Neural Networks
• Deep learning predictions of galaxy merger stage and the importance of observational realism
• Incidence of WISE-Selected Obscured AGNs in Major Mergers and Interactions from the SDSS
• Galaxy Zoo: Quantitative Visual Morphological Classifications for 48,000 galaxies from CANDELS
• Finding high-redshift strong lenses in DES using convolutional neural networks
• Galaxy morphology, luminosity and, environment in the SDSS DR7
• Weak lensing cosmology with convolutional neural networks on noisy data
• Big Universe, Big Data: Machine Learning and Image Analysis for Astronomy
• A Catalog of Edge-on Disk Galaxies: From Galaxies with a Bulge to Superthin Galaxies
• Fast cosmic web simulations with generative adversarial networks
• Galaxy Zoo: Bar Lengths in Nearby Disk Galaxies
• Machine and Deep Learning Applied to Galaxy Morphology -- A Comparative Study
• Transfer learning for galaxy morphology from one survey to another
• Identifying Galaxy Mergers in Observations and Simulations with Deep Learning
• Radio Galaxy Zoo: Compact and extended radio source classification with deep learning
• The different star-formation histories of blue and red spiral and elliptical galaxies
• The Catalog of Edge-on Disk Galaxies from SDSS. Part I: the catalog and the Structural Parameters of Stellar Disks
• VDES J2325-5229 a z=2.7 gravitationally lensed quasar discovered using morphology independent supervised machine learning
• LensExtractor: A Convolutional Neural Network in Search of Strong Gravitational Lenses
• Deep learning Approach for Classifying, Detecting and Predicting Photometric Redshifts of Quasars in the Sloan Digital Sky Survey Stripe 82
• Practical Galaxy Morphology Tools from Deep Supervised Representation Learning
• Data Mining for Gravitationally Lensed Quasars
• Improved $\gamma$/hadron separation for the detection of faint gamma-ray sources using boosted decision trees
• Atlas and Catalog of Collisional Ring Galaxies
• A Machine Learning Based Morphological Classification of 14,245 Radio AGNs Selected From The Best-Heckman Sample
• Galaxy Morphology Network: A Convolutional Neural Network Used to Study Morphology and Quenching in $\sim 100,000$ SDSS and $\sim 20,000$ CANDELS Galaxies
• Deblending galaxy superpositions with branched generative adversarial networks
• Testing Convolutional Neural Networks for finding strong gravitational lenses in KiDS
• KiDS-SQuaD II: Machine learning selection of bright extragalactic objects to search for new gravitationally lensed quasars
• DeepMerge II: Building Robust Deep Learning Algorithms for Merging Galaxy Identification Across Domains
• Machine learning technique for morphological classification of galaxies from the SDSS. I. Photometry-based approach
• Pushing automated morphological classifications to their limits with the Dark Energy Survey
• Forging new worlds: high-resolution synthetic galaxies with chained generative adversarial networks
• Fanaroff-Riley classification of radio galaxies using group-equivariant convolutional neural networks
• Evaluating the optical classification of Fermi BCUs using machine learning
• An environmental dependence of the physical and structural properties in the Hydra Cluster galaxies
• Return of the features. Efficient feature selection and interpretation for photometric redshifts
• Relating the structure of dark matter halos to their assembly and environment
• North Ecliptic Pole merging galaxy catalogue
• Cosmic voids detection without density measurements
• Galaxy Morphology Classification using Neural Ordinary Differential Equations
• New exocomets of $\beta$ Pic
• Galaxy And Mass Assembly: Automatic Morphological Classification of Galaxies Using Statistical Learning
• Photometric decomposition of mergers in disk galaxies
• Low Surface Brightness Galaxy catalogue selected from the alpha.40-SDSS DR7 Survey and Tully-Fisher relation
• X-ray AGN in the XMM-LSS galaxy clusters: no evidence of AGN suppression
• Explaining deep learning of galaxy morphology with saliency mapping
• The Vanishing & Appearing Sources during a Century of Observations project: I. USNO objects missing in modern sky surveys and follow-up observations of a "missing star"
• Towards an Understanding of the Massive Red Spiral Galaxy Formation
• Statistical analysis of probability density functions for photometric redshifts through the KiDS-ESO-DR3 galaxies
• A Bayesian Framework for Cosmic String Searches in CMB Maps
• What drives the scatter of local star-forming galaxies in the BPT diagrams? A Machine Learning based analysis
• Revealing the Local Cosmic Web from Galaxies by Deep Learning
• Low-Density Structures in the Local Universe. II. Nearby Cosmic Voids
• A method of immediate detection of objects with a near-zero apparent motion in series of CCD-frames
• Deep Learning for Line Intensity Mapping Observations: Information Extraction from Noisy Maps
• Deep Learning in Searching the Spectroscopic Redshift of Quasars
• Large-scale structures in the $\Lambda$CDM Universe: network analysis and machine learning
• Classifying galaxy spectra at 0.5<z<1 with self-organizing maps
• The Voronoi tessellation method in astronomy
• The luminosity function of ringed galaxies
• Machine-learning computation of distance modulus for local galaxies
• Labeling Bias in Galaxy Morphologies
• Applications of Machine-Learning Algorithms for Infrared Colour Selection of Galactic Wolf-Rayet Stars
• VEXAS: VISTA EXtension to Auxiliary Surveys -- Data Release 2: Machine-learning based classification of sources in the Southern Hemisphere
• Reliable mass calculation in spherical gravitating systems
• Automatic identification of outliers in Hubble Space Telescope galaxy images

Cited by in corpus (3)

• FNet II: Spectral Classification of Quasars, Galaxies, Stars, and broad absorption line (BAL) Quasars
• Visual inspection of potential exocomet transits identified through machine learning and statistical methods
• Milky Way galaxy-analogs and isolated galaxies with bars: environmental density in the Local Volume