CNN Architecture Comparison for Radio Galaxy Classification

Abstract

The morphological classification of radio sources is important to gain a full understanding of galaxy evolution processes and their relation with local environmental properties. Furthermore, the complex nature of the problem, its appeal for citizen scientists and the large data rates generated by existing and upcoming radio telescopes combine to make the morphological classification of radio sources an ideal test case for the application of machine learning techniques. One approach that has shown great promise recently is Convolutional Neural Networks (CNNs). Literature, however, lacks two major things when it comes to CNNs and radio galaxy morphological classification. Firstly, a proper analysis of whether overfitting occurs when training CNNs to perform radio galaxy morphological classification using a small curated training set is needed. Secondly, a good comparative study regarding the practical applicability of the CNN architectures in literature is required. Both of these shortcomings are addressed in this paper. Multiple performance metrics are used for the latter comparative study, such as inference time, model complexity, computational complexity and mean per class accuracy. As part of this study we also investigate the effect that receptive field, stride length and coverage has on recognition performance. For the sake of completeness, we also investigate the recognition performance gains that we can obtain by employing classification ensembles. A ranking system based upon recognition and computational performance is proposed. MCRGNet, Radio Galaxy Zoo and ConvXpress (novel classifier) are the architectures that best balance computational requirements with recognition performance.

20 pages, 11 figures, journal, accepted for publication by Monthly Notices of the Royal Astronomical Society

References in corpus (41)

• Galaxy Zoo : Morphologies derived from visual inspection of galaxies from the Sloan Digital Sky Survey
• Deep Big Simple Neural Nets Excel on Handwritten Digit Recognition
• Rotation-invariant convolutional neural networks for galaxy morphology prediction
• On the fundamental dichotomy in the local radio-AGN population: accretion, evolution, and host galaxy properties
• Galaxy Zoo 2: detailed morphological classifications for 304,122 galaxies from the Sloan Digital Sky Survey
• EMU: Evolutionary Map of the Universe
• Adding Gradient Noise Improves Learning for Very Deep Networks
• Revisiting the Fanaroff-Riley dichotomy and radio-galaxy morphology with the LOFAR Two-Metre Sky Survey (LoTSS)
• Classifying Radio Galaxies with Convolutional Neural Network
• Radio Galaxy Zoo: host galaxies and radio morphologies derived from visual inspection
• The local radio-galaxy population at 20 GHz
• FIRST `Winged' and `X'-shaped Radio Source Candidates
• FRII{\sl{CAT}}: A FIRST catalog of FR~II radio galaxies
• Anticipated Performance of the Square Kilometre Array -- Phase 1 (SKA1)
• FR0CAT: a FIRST catalog of FR0 radio galaxies
• Radio Galaxy Zoo: ClaRAN - A Deep Learning Classifier for Radio Morphologies
• Radio Continuum Surveys with Square Kilometre Array Pathfinders
• FRICAT: A FIRST catalog of FRI radio galaxies
• GAMA/WiggleZ: The 1.4GHz radio luminosity functions of high- and low-excitation radio galaxies and their redshift evolution to z=0.75
• Radio Galaxy Zoo: Compact and extended radio source classification with deep learning
• The nuclear properties and extended morphologies of powerful radio galaxies: the roles of host galaxy and environment
• Hydrodynamical Backflow in X-shaped Radio Galaxy PKS 2014-55
• Morphological Annotations for Groups in the FIRST Database
• Morphological classification of radio galaxies: Capsule Networks versus Convolutional Neural Networks
• A Machine Learning Based Morphological Classification of 14,245 Radio AGNs Selected From The Best-Heckman Sample
• DeepSource: Point Source Detection using Deep Learning
• Radio Galaxy Zoo: Machine learning for radio source host galaxy cross-identification
• The FIRST Classifier: Compact and Extended Radio Galaxy Classification using Deep Convolutional Neural Networks
• Unveiling the cause of hybrid morphology radio sources (HyMoRS)
• The CoNFIG Catalogue - II. Comparison of Space Densities in the FR Dichotomy
• WATCAT: a tale of wide-angle tailed radio galaxies
• Transfer learning for radio galaxy classification
• Deep Learning Based Detection of Cosmological Diffuse Radio Sources
• Deep GMRT 610 MHz Observations of the ELAIS N1 Field : Catalogue and Source Counts
• Radio sources associated with Optical Galaxies and having Unresolved or Extended morphologies (ROGUE). I. A catalog of SDSS galaxies with FIRST core identifications
• FR0 radio galaxies and their place in the radio morphology classification
• The Stripe 82 1-2 GHz Very Large Array Snapshot Survey: Multiwavelength Counterparts
• The Combined NVSS-FIRST Galaxies (CoNFIG) Sample - I. Sample Definition, Classification and Evolution
• What Are "X-Shaped" Radio Sources Telling Us? I. Very Large Array Imaging of a Large Sample of Low Axial Ratio Radio Sources
• The morphological classification of distant radio galaxies explored with three-dimensional simulations
• The faint radio source population at 15.7 GHz -- IV. The dominance of core emission in faint radio galaxies

Cited by in corpus (18)

• Improving the open cluster census. II. An all-sky cluster catalogue with Gaia DR3
• The DAWES review 10: The impact of deep learning for the analysis of galaxy surveys
• Life cycle of cosmic-ray electrons in the intracluster medium
• Artificial intelligence for celestial object census: the latest technology meets the oldest science
• A Multimodal Dataset and Benchmark for Radio Galaxy and Infrared Host Detection
• Denoising Diffusion Probabilistic Models to Predict the Density of Molecular Clouds
• Feature Guided Training and Rotational Standardisation for the Morphological Classification of Radio Galaxies
• A Comparison of Deep Learning Architectures for Optical Galaxy Morphology Classification
• Identification of 4876 Bent-Tail Radio Galaxies in the FIRST Survey Using Deep Learning Combined with Visual Inspection
• The Classification of Optical Galaxy Morphology Using Unsupervised Learning Techniques
• Morphological Classification of Extragalactic Radio Sources Using Gradient Boosting Methods
• Reducing false positives in strong lens detection through effective augmentation and ensemble learning
• Radio Galaxy Zoo: Using semi-supervised learning to leverage large unlabelled data-sets for radio galaxy classification under data-set shift
• E(2)-Equivariant Features in Machine Learning for Morphological Classification of Radio Galaxies
• Lenses In VoicE (LIVE): Searching for strong gravitational lenses in the VOICE@VST survey using Convolutional Neural Networks
• Radio Galaxy Zoo: Giant Radio Galaxy Classification using Multi-Domain Deep Learning
• Quantifying Radio Source Morphology
• Can semi-supervised learning reduce the amount of manual labelling required for effective radio galaxy morphology classification?