Square Kilometre Array Science Data Challenge 1: analysis and results

Abstract

As the largest radio telescope in the world, the Square Kilometre Array (SKA) will lead the next generation of radio astronomy. The feats of engineering required to construct the telescope array will be matched only by the techniques developed to exploit the rich scientific value of the data. To drive forward the development of efficient and accurate analysis methods, we are designing a series of data challenges that will provide the scientific community with high-quality datasets for testing and evaluating new techniques. In this paper we present a description and results from the first such Science Data Challenge (SDC1). Based on SKA MID continuum simulated observations and covering three frequencies (560 MHz, 1400MHz and 9200 MHz) at three depths (8 h, 100 h and 1000 h), SDC1 asked participants to apply source detection, characterization and classification methods to simulated data. The challenge opened in November 2018, with nine teams submitting results by the deadline of April 2019. In this work we analyse the results for 8 of those teams, showcasing the variety of approaches that can be successfully used to find, characterise and classify sources in a deep, crowded field. The results also demonstrate the importance of building domain knowledge and expertise on this kind of analysis to obtain the best performance. As high-resolution observations begin revealing the true complexity of the sky, one of the outstanding challenges emerging from this analysis is the ability to deal with highly resolved and complex sources as effectively as the unresolved source population.

arXiv comment18 pages, 11 figures, accepted by MNRAS

References in corpus (29)

• SciPy 1.0--Fundamental Algorithms for Scientific Computing in Python
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Astropy: A Community Python Package for Astronomy
• Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising
• The Wide-field Infrared Survey Explorer (WISE): Mission Description and Initial On-orbit Performance
• The Astropy Project: Building an inclusive, open-science project and status of the v2.0 core package
• scikit-image: Image processing in Python
• Deep Residual Learning for Image Recognition
• Feature Pyramid Networks for Object Detection
• Compact continuum source-finding for next generation radio surveys
• Resolving the Radio Source Background: Deeper Understanding Through Confusion
• ProFound: Source Extraction and Application to Modern Survey Data
• Source finding in the era of the SKA (precursors): Aegean 2.0
• Source extraction and photometry for the far-infrared and sub-millimeter continuum in the presence of complex backgrounds
• A new source detection algorithm using FDR
• Radio Galaxy Zoo: host galaxies and radio morphologies derived from visual inspection
• The Tiered Radio Extragalactic Continuum Simulation (T-RECS)
• BLOBCAT: Software to Catalogue Flood-Filled Blobs in Radio Images of Total Intensity and Linear Polarization
• Duchamp: a 3D source finder for spectral-line data
• Radio Galaxy Zoo: ClaRAN - A Deep Learning Classifier for Radio Morphologies
• Source-finding for the Australian Square Kilometre Array Pathfinder
• DeepSource: Point Source Detection using Deep Learning
• The ASKAP/EMU Source Finding Data Challenge
• Radio Galaxy Zoo: A Search for Hybrid Morphology Radio Galaxies
• Radio Source Extraction with ProFound
• Deep 3 GHz Observations of the Lockman Hole North with the Very Large Array - 1. Source extraction and uncertainty analysis
• Radio Galaxy Zoo: discovery of a poor cluster through a giant wide-angle tail radio galaxy
• SuperCLASS -- III. Weak lensing from radio and optical observations in Data Release 1
• A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection

Cited by in corpus (22)

• Deep Radio Interferometric Imaging with POLISH: DSA-2000 and weak lensing
• Status and progress of China SKA Regional Centre prototype
• SKA Science Data Challenge 2: analysis and results
• Artificial intelligence for celestial object census: the latest technology meets the oldest science
• Radio astronomical images object detection and segmentation: A benchmark on deep learning methods
• Intermediate-mass black holes: finding of episodic, large-scale and powerful jet activity in a dwarf galaxy
• YOLO-CIANNA: Galaxy detection with deep learning in radio data. I. A new YOLO-inspired source detection method applied to the SKAO SDC1
• The Tiered Radio Extragalactic Continuum (T-RECS) simulation II: HI emission and continuum-HI cross-correlation
• Radio source-component association for the LOFAR Two-metre Sky Survey with region-based convolutional neural networks
• Morphological Classification of Radio Galaxies with wGAN-supported Augmentation
• Dark Matter Fraction in z~1 Star-Forming Galaxies
• Selection of powerful radio galaxies with machine learning
• The Effect of Light Deflection by Solar System Objects on High-Precision SKA Astrometry
• Lightweight HI source finding for next generation radio surveys
• Simulating images of radio galaxies with diffusion models
• Transforming a rare event search into a not-so-rare event search in real-time with deep learning-based object detection
• Bringing computation to the data: A MOEA-driven approach for optimising data processing in the context of the SKA and SRCNet
• Detecting galaxies in a large H{\sc i}~spectral cube
• An empirical model of the extragalactic radio background
• Transformer-Based Source Detection and Morphological Classification in LOFAR Deep-Field Continuum Images
• Light Deflection under the Gravitational Field of Jupiter -- Testing General Relativity
• A Cloud-native Agile approach to cyber platform prototyping and integration for astronomy: the ENGAGE SKA case