Testing Convolutional Neural Networks for finding strong gravitational lenses in KiDS
C. E. Petrillo, C. Tortora, S. Chatterjee, G. Vernardos, L. V. E. Koopmans, G. Verdoes Kleijn, N. R. Napolitano, G. Covone, L. S. Kelvin, A. M. Hopkins
备注：
16 pages, 10 figures. Accepted for publication in MNRAS
Convolutional Neural Networks (ConvNets) are one of the most promising methods for identifying strong gravitational lens candidates in survey data. We present two ConvNet lens-finders which we have trained with a dataset composed of real galaxies from the Kilo Degree Survey (KiDS) and simulated lensed sources. One ConvNet is trained with single \textit{r}-band galaxy images, hence basing the classification mostly on the morphology. While the other ConvNet is trained on \textit{g-r-i} composite images, relying mostly on colours and morphology. We have tested the ConvNet lens-finders on a sample of 21789 Luminous Red Galaxies (LRGs) selected from KiDS and we have analyzed and compared the results with our previous ConvNet lens-finder on the same sample. The new lens-finders achieve a higher accuracy and completeness in identifying gravitational lens candidates, especially the single-band ConvNet. Our analysis indicates that this is mainly due to improved simulations of the lensed sources. In particular, the single-band ConvNet can select a sample of lens candidates with $\sim40\%$ purity, retrieving 3 out of 4 of the confirmed gravitational lenses in the LRG sample. With this particular setup and limited human intervention, it will be possible to retrieve, in future surveys such as Euclid, a sample of lenses exceeding in size the total number of currently known gravitational lenses.
New methods have been recently developed to search for strong gravitational lenses, in particular lensed quasars, in wide-field imaging surveys. Here, we compare the performance of three different, morphology- and photometry- based methods to find lens candidates over the Kilo-Degree Survey (KiDS) DR3 footprint (440 deg$^2$). The three methods are: i) a multiplet detection in KiDS-DR3 and/or Gaia-DR1, ii) direct modeling of KiDS cutouts and iii) positional offsets between different surveys (KiDS-vs-Gaia, Gaia-vs-2MASS), with purpose-built astrometric recalibrations. The first benchmark for the…
We report the discovery of a candidate to quadrupole gravitationally lensed system KiDS0239-3211 based on the public data release 3 of the KiDS survey and machine learning techniques.
Identification of gravitational lenses obscured by foreground light in the KiDS dataset using U-Nets and ResNets
Wuhan University（武汉大学）；Zhengzhou University（郑州大学）；Nanchang University（南昌大学）；INAF – Osservatorio Astronomico di Capodimonte（意大利国家天体物理研究所卡波迪蒙特天文台）；University Federico II（那不勒斯费德里科二世大学）；Sun Yat-sen University（中山大学）；Yunnan Observatories, Chinese Academy of Sciences（中国科学院云南天文台）；Xi’an Jiaotong-Liverpool University（西交利物浦大学）；Beijing Normal University（北京师范大学）
Despite the success of galaxy-scale strong gravitational lens studies with Hubble-quality imaging, the number of well-studied strong lenses remains small. As a result, robust comparisons of the lens models to theoretical predictions are difficult. This motivates our application of automated Bayesian lens modeling methods to observations from public data releases of overlapping large ground-based imaging and spectroscopic surveys: Kilo-Degree Survey (KiDS) and Galaxy and Mass Assembly (GAMA), respectively. We use the open-source lens modeling software PyAutoLens to perform our analysis. We demo…
The volume of data that will be produced by new-generation surveys requires automatic classification methods to select and analyze sources. Indeed, this is the case for the search for strong gravitational lenses, where the population of the detectable lensed sources is only a very small fraction of the full source population. We apply for the first time a morphological classification method based on a Convolutional Neural Network (CNN) for recognizing strong gravitational lenses in $255$ square degrees of the Kilo Degree Survey (KiDS), one of the current-generation optical wide surveys. The CN…
In this work, we present our classification algorithm to identify strong gravitational lenses from wide-area surveys using machine learning convolutional neural network; LensExtractor. We train and test the algorithm using a wide variety of strong gravitational lens configurations from simulations of lensing events. Images are processed through multiple convolutional layers which extract feature maps necessary to assign a lens probability to each image. LensExtractor provides a ranking scheme for all sources which could be used to identify potential gravitational lens candidates significantly …
The Euclid telescope, due for launch in 2021, will perform an imaging and slitless spectroscopy survey over half the sky, to map baryon wiggles and weak lensing. During the survey Euclid is expected to resolve 100,000 strong gravitational lens systems. This is ideal to find rare lens configurations, provided they can be identified reliably and on a reasonable timescale. For this reason we have developed a Convolutional Neural Network (CNN) that can be used to identify images containing lensing systems. CNNs have already been used for image and digit classification as well as being used in astr…
We train and apply convolutional neural networks, a machine learning technique developed to learn from and classify image data, to Canada-France-Hawaii Telescope Legacy Survey (CFHTLS) imaging for the identification of potential strong lensing systems. An ensemble of four convolutional neural networks was trained on images of simulated galaxy-galaxy lenses. The training sets consisted of a total of 62,406 simulated lenses and 64,673 non-lens negative examples generated with two different methodologies. The networks were able to learn the features of simulated lenses with accuracy of up to 99.8…
We have started a systematic search of strong lens candidates in the ESO public survey KiDS based on the visual inspection of massive galaxies in the redshift range $0.1<z<0.5$. As a pilot program we have inspected 100 sq. deg., which overlap with SDSS and where there are known lenses to use as a control sample. Taking advantage of the superb image quality of VST/OmegaCAM, the colour information and accurate model subtracted images, we have found 18 new lens candidates, for which spectroscopic confirmation will be needed to confirm their lensing nature and study the mass profile of the lensing…
Gravitationally lensed quasars are valuable, but extremely rare, probes of observational cosmology and extragalactic astrophysics. Progress in these fields has been limited just by the paucity of systems with good ancillary data. Here we present a first spectroscopic confirmation of lenses discovered in the Southern Sky from the DES and KiDS-DR3 footprints. % optical ground based surveys in the Southern Hemisphere. We have targeted 7 high-graded candidates, selected with new techniques, with NTT-EFOSC2, and confirmed 5 of them. We provide source spectroscopic redshifts, image separations, $gri…
Strong gravitational lenses are a rare and instructive type of astronomical object. Identification has long relied on serendipity, but different strategies -- such as mixed spectroscopy of multiple galaxies along the line of sight, machine learning algorithms, and citizen science -- have been employed to identify these objects as new imaging surveys become available. We report on the comparison between spectroscopic, machine learning, and citizen science identification of galaxy-galaxy lens candidates from independently constructed lens catalogs in the common survey area of the equatorial fiel…
We present a sample of 16 likely strong gravitational lenses identified in the VST Optical Imaging of the CDFS and ES1 fields (VOICE survey) using Convolutional Neural Networks (CNNs). We train two different CNNs on composite images produced by superimposing simulated gravitational arcs on real Luminous Red Galaxies observed in VOICE. Specifically, the first CNN is trained on single-band images and more easily identifies systems with large Einstein radii, while the second one, trained on composite RGB images, is more accurate in retrieving systems with smaller Einstein radii. We apply both net…
We search Dark Energy Survey (DES) Year 3 imaging data for galaxy-galaxy strong gravitational lenses using convolutional neural networks. We generate 250,000 simulated lenses at redshifts > 0.8 from which we create a data set for training the neural networks with realistic seeing, sky and shot noise. Using the simulations as a guide, we build a catalogue of 1.1 million DES sources with 1.8 < g - i < 5, 0.6 < g -r < 3, r_mag > 19, g_mag > 20 and i_mag > 18.2. We train two ensembles of neural networks on training sets consisting of simulated lenses, simulated non-lenses, and real sources. We use…
From DES to KiDS: Domain adaptation for cross-survey detection of low-surface-brightness galaxies
National Centre for Nuclear Research（国家核研究中心）；Universidad de Valladolid（巴利亚多利德大学）；Laboratory for Disruptive Interdisciplinary Science (LaDIS)（颠覆性跨学科科学实验室）；Instituto de Astrofísica de Canarias（加那利天体物理研究所）；Universidad de La Laguna（拉古纳大学）
In this paper we develop a new unsupervised machine learning technique comprised of a feature extractor, a convolutional autoencoder (CAE), and a clustering algorithm consisting of a Bayesian Gaussian mixture model (BGM). We apply this technique to visual band space-based simulated imaging data from the Euclid Space Telescope using data from the Strong Gravitational Lenses Finding Challenge. Our technique promisingly captures a variety of lensing features such as Einstein rings with different radii, distorted arc structures, etc, without using predefined labels. After the clustering process, w…
We present a new sample of galaxy-scale strong gravitational-lens candidates, selected from 904 square degrees of Data Release 4 of the Kilo-Degree Survey (KiDS), i.e., the "Lenses in the Kilo-Degree Survey" (LinKS) sample. We apply two Convolutional Neural Networks (ConvNets) to $\sim88\,000$ colour-magnitude selected luminous red galaxies yielding a list of 3500 strong-lens candidates. This list is further down-selected via human inspection. The resulting LinKS sample is composed of 1983 rank-ordered targets classified as "potential lens candidates" by at least one inspector. Of these, a hig…
We present spectroscopic confirmation of candidate strong gravitational lenses using the Keck Observatory and Very Large Telescope as part of our ASTRO 3D Galaxy Evolution with Lenses (AGEL) survey. We confirm that 1) search methods using Convolutional Neural Networks (CNN) with visual inspection successfully identify strong gravitational lenses and 2) the lenses are at higher redshifts relative to existing surveys due to the combination of deeper and higher resolution imaging from DECam and spectroscopy spanning optical to near-infrared wavelengths. We measure 104 redshifts in 77 systems sele…
Confirming lensed-quasar candidates with DESI and P200 spectroscopy I. 14 lensed quasars and 8 lensed galaxies
Nanchang University（南昌大学）；Purple Mountain Observatory, Chinese Academy of Sciences（中国科学院紫金山天文台）；Beijing Normal University（北京师范大学）；University of Hawaiʻi（夏威夷大学）；University of San Francisco（旧金山大学）；National Astronomical Observatories（国家天文台）；University of Science and Technology of China（中国科学技术大学）
Forthcoming large imaging surveys such as Euclid and the Vera Rubin Observatory Legacy Survey of Space and Time are expected to find more than $10^5$ strong gravitational lens systems, including many rare and exotic populations such as compound lenses, but these $10^5$ systems will be interspersed among much larger catalogues of $\sim10^9$ galaxies. This volume of data is too much for visual inspection by volunteers alone to be feasible and gravitational lenses will only appear in a small fraction of these data which could cause a large amount of false positives. Machine learning is the obviou…
We search Dark Energy Survey (DES) Year 3 imaging for galaxy-galaxy strong gravitational lenses using convolutional neural networks, extending previous work with new training sets and covering a wider range of redshifts and colors. We train two neural networks using images of simulated lenses, then use them to score postage stamp images of 7.9 million sources from the Dark Energy Survey chosen to have plausible lens colors based on simulations. We examine 1175 of the highest-scored candidates and identify 152 probable or definite lenses. Examining an additional 20,000 images with lower scores,…
Reducing False Positives in Strong-Lens Searches with Generalized-Mean Consensus of Machine-Learning Ensembles in the Kilo-Degree Survey
Institute for Astrophysics, School of Physics, Zhengzhou University（郑州大学物理学院天体物理研究所）；International College, Zhengzhou University（郑州大学国际教育学院）；School of Physics and Astronomy, Beijing Normal University（北京师范大学天文与物理学院）；INAF – Osservatorio Astronomico di Capodimonte（意大利国家天体物理研究所卡波迪蒙特天文台）；Department of Physics “E. Pancini”, University Federico II（那不勒斯费德里科二世大学E. Pancini物理系）；School of Mathematics and Physics, Xi’an Jiaotong-Liverpool University（西交利物浦大学数理学院）
We present a systematic search for wide-separation (Einstein radius >1.5"), galaxy-scale strong lenses in the 30 000 sq.deg of the Pan-STARRS 3pi survey on the Northern sky. With long time delays of a few days to weeks, such systems are particularly well suited for catching strongly lensed supernovae with spatially-resolved multiple images and open new perspectives on early-phase supernova spectroscopy and cosmography. We produce a set of realistic simulations by painting lensed COSMOS sources on Pan-STARRS image cutouts of lens luminous red galaxies with known redshift and velocity dispersion…
We present a bright galaxy sample with accurate and precise photometric redshifts (photo-zs), selected using $ugriZYJHK_\mathrm{s}$ photometry from the Kilo-Degree Survey (KiDS) Data Release 4 (DR4). The highly pure and complete dataset is flux-limited at $r<20$ mag, covers $\sim1000$ deg$^2$, and contains about 1 million galaxies after artifact masking. We exploit the overlap with Galaxy And Mass Assembly (GAMA) spectroscopy as calibration to determine photo-zs with the supervised machine learning neural network algorithm implemented in the ANNz2 software. The photo-zs have mean error of $|\l…
LenNet: Direct Detection and Localization of Strong Gravitational Lenses in Wide-Field Sky Survey Images
Institute for Astrophysics, School of Physics, Zhengzhou University（郑州大学物理学院天体物理研究所）；International College, Zhengzhou University（郑州大学国际教育学院）；School of Astronomy and Space Science, University of Chinese Academy of Sciences（中国科学院大学天文与空间科学学院）；National Astronomical Observatories, Chinese Academy of Sciences（中国科学院国家天文台）；Department of Physics “E. Pancini”, University of Naples Federico II（那不勒斯费德里科二世大学E. Pancini物理系）；School of Physics and Astronomy, Beijing Normal University（北京师范大学物理与天文学院）；Department of Physics “E. Pancini”, University Federico II（费德里科二世大学E. Pancini物理系）；Kapteyn Astronomical Institute, University of Groningen（格罗宁根大学卡普坦天文研究所）；INAF – Osservatorio Astronomico di Capodimonte（意大利国家天体物理研究所卡波迪蒙特天文台）
University of Groningen（格罗宁根大学）；ASTRON, Institute for Radio Astronomy（荷兰射电天文研究所）；University of Birmingham（伯明翰大学）
We present a novel machine learning based approach for detecting galaxy-scale gravitational lenses from interferometric data, specifically those taken with the International LOFAR Telescope (ILT), which is observing the northern radio sky at a frequency of 150 MHz, an angular resolution of 350 mas and a sensitivity of 90 uJy beam-1 (1 sigma). We develop and test several Convolutional Neural Networks to determine the probability and uncertainty of a given sample being classified as a lensed or non-lensed event. By training and testing on a simulated interferometric imaging data set that include…
The Kilo-Degree Survey (KiDS) is an optical wide-field imaging survey carried out with the VLT Survey Telescope and the OmegaCAM camera. KiDS will image 1500 square degrees in four filters (ugri), and together with its near-infrared counterpart VIKING will produce deep photometry in nine bands. Designed for weak lensing shape and photometric redshift measurements, the core science driver of the survey is mapping the large-scale matter distribution in the Universe back to a redshift of ~0.5. Secondary science cases are manifold, covering topics such as galaxy evolution, Milky Way structure, and…
We present the catalogue of blended galaxy spectra from the Galaxy And Mass Assembly (GAMA) survey. These are cases where light from two galaxies are significantly detected in a single GAMA fibre. Galaxy pairs identified from their blended spectrum fall into two principal classes: they are either strong lenses, a passive galaxy lensing an emission-line galaxy; or occulting galaxies, serendipitous overlaps of two galaxies, of any type. Blended spectra can thus be used to reliably identify strong lenses for follow-up observations (high resolution imaging) and occulting pairs, especially those th…
Next generation large sky surveys will observe up to billions of galaxies for which basic structural parameters are needed to study their evolution. This is a challenging task that, for ground-based observations, is complicated by seeing limited point-spread function (PSF). To perform a fast and accurate analysis of galaxy surface brightness, we have developed a family of supervised Convolutional Neural Networks (CNN) to derive Sersic profile parameters of galaxies. This work presents the first two Galaxy Light profile convolutional neural Networks (GaLNets) of this family. The first one is tr…
Reducing false positives in strong lens detection through effective augmentation and ensemble learning
Leiden Observatory, Leiden University（莱顿大学莱顿天文台）；Leiden Institute of Advanced Computer Science (LIACS), Leiden University（莱顿大学高级计算机科学研究所）；INFN-Padova（帕多瓦国家核物理研究所）；Kapteyn Astronomical Institute, University of Groningen（格罗宁根大学卡普坦天文研究所）；South African Radio Astronomy Observatory (SARAO)（南非射电天文观测台）；Department of Physics, University of Pretoria（比勒陀利亚大学物理系）
Large scale imaging surveys will increase the number of galaxy-scale strong lensing candidates by maybe three orders of magnitudes beyond the number known today. Finding these rare objects will require picking them out of at least tens of millions of images and deriving scientific results from them will require quantifying the efficiency and bias of any search method. To achieve these objectives automated methods must be developed. Because gravitational lenses are rare objects reducing false positives will be particularly important. We present a description and results of an open gravitational…
Future large-scale surveys with high resolution imaging will provide us with a few $10^5$ new strong galaxy-scale lenses. These strong lensing systems however will be contained in large data amounts which are beyond the capacity of human experts to visually classify in a unbiased way. We present a new strong gravitational lens finder based on convolutional neural networks (CNNs). The method was applied to the Strong Lensing challenge organised by the Bologna Lens Factory. It achieved first and third place respectively on the space-based data-set and the ground-based data-set. The goal was to f…
We present 97 new high-quality strong lensing candidates found in the final $\sim 350\,\rm deg^2$, that completed the full $\sim 1350\,\rm deg^2$ area of the Kilo-Degree Survey (KiDS). Together with our previous findings, the final list of high-quality candidates from KiDS sums up to 268 systems. The new sample is assembled using a new Convolutional Neural Network (CNN) classifier applied to $r$-band (best seeing) and $g,~r,~i$ color-composited images separately. This optimizes the complementarity of the morphology and color information on the identification of strong lensing candidates. We ap…
Streamlined Lensed Quasar Identification in Multiband Images via Ensemble Networks
Max-Planck-Institut für Astrophysik（马克斯·普朗克天体物理研究所）；Academia Sinica Institute of Astronomy and Astrophysics (ASIAA)（中央研究院天文及天文物理研究所）；Università degli Studi di Milano（米兰大学）；Purple Mountain Observatory（紫金山天文台）；MIT Kavli Institute for Astrophysics and Space Research（麻省理工学院卡夫利天体物理与空间研究所）；Institut Teknologi Bandung（万隆理工学院）；Technical University of Munich（慕尼黑工业大学）
We present a search for galaxy-scale strong gravitational lenses in the initial 2 500 square degrees of the Canada-France Imaging Survey (CFIS). We designed a convolutional neural network (CNN) committee that we applied to a selection of 2 344 002 exquisite-seeing $r$-band images of color-selected luminous red galaxies (LRGs). Our classification uses a realistic training set where the lensing galaxies and the lensed sources are both taken from real data, namely the CFIS $r$-band images themselves and the Hubble Space Telescope (HST). A total of 9 460 candidates obtain a score above 0.5 with th…
We performed a search for strong lens galaxy-scale systems in the first data release of the Dark Energy Survey (DES), from a color-selected parent sample of 18~745~029 Luminous Red Galaxies (LRGs). Our search was based on a Convolutional Neural Network (CNN) to grade our LRG selection with values between 0 (non-lens) and 1 (lens). Our training set was data-driven, i.e. using lensed sources taken from HST COSMOS images and where the light distribution of the lens plane was taken directly from DES images of our LRGs. A total of 76~582 cutouts obtained a score above 0.9. These were visually inspe…
With the advent of next-generation surveys and the expectation of discovering huge numbers of strong gravitational lens systems, much effort is being invested into developing automated procedures for handling the data. The several orders of magnitude increase in the number of strong galaxy-galaxy lens systems is an insurmountable challenge for traditional modelling techniques. Whilst machine learning techniques have dramatically improved the efficiency of lens modelling, parametric modelling of the lens mass profile remains an important tool for dealing with complex lensing systems. In particu…
The vast quantity of strong galaxy-galaxy gravitational lenses expected by future large-scale surveys necessitates the development of automated methods to efficiently model their mass profiles. For this purpose, we train an approximate Bayesian convolutional neural network (CNN) to predict mass profile parameters and associated uncertainties, and compare its accuracy to that of conventional parametric modelling for a range of increasingly complex lensing systems. These include standard smooth parametric density profiles, hydrodynamical EAGLE galaxies and the inclusion of foreground mass struct…
GraViT: Transfer Learning with Vision Transformers and MLP-Mixer for Strong Gravitational Lens Discovery
Facultad de Ingeniería, Universidad Autónoma de San Luis Potosí（圣路易斯波托西自治大学工程学院）；Universidad Nacional Autónoma de México. Instituto de Astronomía（墨西哥国家自治大学天文研究所）；Inter-University Centre for Astronomy and Astrophysics (IUCAA)（天文学与天体物理学跨大学研究中心）；Kavli IPMU (WPI), UTIAS, The University of Tokyo（Kavli IPMU（WPI），东京大学UTIAS）；Astronomy Research Group and Bosscha Observatory, FMIPA, Institut Teknologi Bandung（天文学研究组和博斯卡天文台，万隆理工学院）；U-CoE AI-VLB, Institut Teknologi Bandung（人工智能与视觉学习卓越中心，万隆理工学院）；University Center of Excellence for Space Science, Technology and Innovation, Institut Teknologi Bandung（空间科学、技术和创新卓越大学中心，万隆理工学院）
Searching for strong lensing by late-type galaxies in UNIONS
Ecole Polytechnique Fédérale de Lausanne (EPFL)（洛桑联邦理工学院）；Institut de Ciències del Cosmos (ICCUB), Universitat de Barcelona（巴塞罗那大学宇宙科学研究所）；Institució Catalana de Recerca i Estudis Avançats (ICREA)（加泰罗尼亚高级研究学院）；Aix-Marseille Université（艾克斯-马赛大学）；Sorbonne Université（索邦大学）；Stockholm University（斯德哥尔摩大学）；University of Applied Sciences and Arts of Northwestern Switzerland（瑞士西北应用科学与艺术大学）
Euclid Quick Data Release (Q1). Searching for giant gravitational arcs in galaxy clusters with mask region-based convolutional neural networks
Università degli Studi di Ferrara（费拉拉大学）；INAF-Osservatorio di Astrofisica e Scienza dello Spazio di Bologna（博洛尼亚天文物理与空间科学研究所）；INAF-Osservatorio Astronomico di Capodimonte（卡波迪蒙特天文台）；Dipartimento di Fisica "Aldo Pontremoli", Università degli Studi di Milano（米兰大学阿尔多·蓬特罗利物理系）；INFN-Sezione di Bologna（博洛尼亚意大利国家核物理研究所）；Instituto de Física de Cantabria（坎塔布里亚物理研究所）；INAF-IASF Milano（米兰天文物理与空间科学研究所）
