Classifying Radio Galaxies with the Convolutional Neural Network

We present the application of deep machine learning technique to classify radio images of extended sources on a morphological basis using convolutional neural networks. In this study, we have taken the case of Fanaroff-Riley (FR) class of radio galaxies as well as radio galaxies with bent-tailed morphology. We have used archival data from the Very Large Array (VLA) - Faint Images of the Radio Sky at Twenty Centimeters (FIRST) survey and existing visually classified samples available in literature to train a neural network for morphological classification of these categories of radio sources. Our training sample size for each of these categories is approximately 200 sources, which has been augmented by rotated versions of the same. Our study shows that convolutional neural networks can classify images of the FRI and FRII and bent-tailed radio galaxies with high accuracy (maximum precision at 95%) using well-defined samples and fusion classifier, which combines the results of binary classifications, while allowing for a mechanism to find sources with unusual morphologies. The individual precision is highest for bent-tailed radio galaxies at 95% and is 91% and 75% for the FRI and FRII classes, respectively, whereas the recall is highest for FRI and FRIIs at 91% each, while bent-tailed class has a recall of 79%. These results show that our results are comparable to that of manual classification while being much faster. Finally, we discuss the computational and data-related challenges associated with morphological classification of radio galaxies with convolutional neural networks.

[1]  David G. Stork,et al.  Pattern Classification , 1973 .

[2]  J. Riley,et al.  The Morphology of Extragalactic Radio Sources of High and Low Luminosity , 1974 .

[3]  Hecht-Nielsen.  Theory of the backpropagation neural network , 1989 .

[4]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1992, Math. Control. Signals Syst..

[5]  E. Greisen,et al.  The NRAO VLA Sky Survey , 1996 .

[6]  Hsinchun Chen.  Machine learning for information retrieval: neural networks, symbolic learning, and genetic algorithms , 1995 .

[7]  Richard L. White,et al.  The FIRST Survey: Faint Images of the Radio Sky at twenty centimeters , 1995 .

[8]  S. Djorgovski,et al.  Automated Star/Galaxy Classification for Digitized Poss-II , 1995 .

[9]  Anil K. Jain,et al.  Artificial Neural Networks: A Tutorial , 1996, Computer.

[10]  Frazer N. Owen,et al.  A 20cm VLA Survey of Abell Clusters of Galaxies VI. Radio/Optical Luminosity Functions , 1996 .

[11]  Joachim Hagenauer,et al.  Iterative decoding of binary block and convolutional codes , 1996, IEEE Trans. Inf. Theory.

[12]  Andrew P. Bradley,et al.  The use of the area under the ROC curve in the evaluation of machine learning algorithms , 1997, Pattern Recognit..

[13]  Ah Chung Tsoi,et al.  Face recognition: a convolutional neural-network approach , 1997, IEEE Trans. Neural Networks.

[14]  Pat Langley,et al.  Selection of Relevant Features and Examples in Machine Learning , 1997, Artif. Intell..

[15]  Burns.  Stormy weather in galaxy clusters , 1997, Science.

[16]  N. Benı́tez.  Bayesian Photometric Redshift Estimation , 1998, astro-ph/9811189.

[17]  David G. Stork,et al.  Pattern Classification (2nd ed.) , 1999 .

[18]  Deanne D. Proctor,et al.  Low-resolution pattern recognition - sorting triples in the FIRST database , 2003, J. Electronic Imaging.

[19]  Volker Springel,et al.  The Many lives of AGN: Cooling flows, black holes and the luminosities and colours of galaxies , 2006, astro-ph/0602065.

[20]  Shie Mannor,et al.  A Tutorial on the Cross-Entropy Method , 2005, Ann. Oper. Res..

[21]  Isabelle Guyon,et al.  An Introduction to Feature Extraction , 2006, Feature Extraction.

[22]  G. Kauffmann,et al.  The many lives of active galactic nuclei: cooling flows, black holes and the luminosities and colour , 2005, astro-ph/0508046.

[23]  Geoffrey E. Hinton,et al.  Reducing the Dimensionality of Data with Neural Networks , 2006, Science.

[24]  D. Proctor.  Comparing Pattern Recognition Feature Sets for Sorting Triples in the FIRST Database , 2006, astro-ph/0605104.

[25]  Yee Whye Teh,et al.  A Fast Learning Algorithm for Deep Belief Nets , 2006, Neural Computation.

[26]  S. Kotsiantis.  Supervised Machine Learning: A Review of Classification Techniques , 2007, Informatica.

[27]  Jason Weston,et al.  Large-Scale Kernel Machines (Neural Information Processing) , 2007 .

[28]  J. Wall,et al.  The Combined NVSS–FIRST Galaxies (CoNFIG) sample – I. Sample definition, classification and evolution , 2008, 0808.0165.

[29]  Geoffrey I. Webb,et al.  Encyclopedia of Machine Learning , 2011, Encyclopedia of Machine Learning.

[30]  M. Jarvis,et al.  An infrared–radio simulation of the extragalactic sky: from the Square Kilometre Array to Herschel , 2010, 1002.1112.

[31]  R. P. Eatough,et al.  Selection of radio pulsar candidates using artificial neural networks , 2010, 1005.5068.

[32]  P. Best,et al.  The CoNFIG Catalogue - II. Comparison of Space Densities in the FR Dichotomy , 2010, 1001.4514.

[33]  Derek C. Rose,et al.  Deep Machine Learning - A New Frontier in Artificial Intelligence Research [Research Frontier] , 2010, IEEE Computational Intelligence Magazine.

[34]  P. Best,et al.  The Combined NVSS–FIRST Galaxies (CoNFIG) sample – II. Comparison of space densities in the Fanaroff–Riley dichotomy , 2010 .

[35]  D. J. Saikia,et al.  EMU: Evolutionary Map of the Universe , 2011, Publications of the Astronomical Society of Australia.

[36]  D. Proctor.  MORPHOLOGICAL ANNOTATIONS FOR GROUPS IN THE FIRST DATABASE , 2011, 1104.3896.

[37]  D. J. Saikia,et al.  ATLAS, and Wide-Angle Tail Galaxies in ATLAS , 2011 .

[38]  S. G. Djorgovski,et al.  Discovery, classification, and scientific exploration of transient events from the Catalina Real-time Transient Survey , 2011, 1111.0313.

[39]  Tara N. Sainath,et al.  Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research Groups , 2012, IEEE Signal Processing Magazine.

[40]  Nikolaos Papanikolopoulos,et al.  Scalable Active Learning for Multiclass Image Classification , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[41]  S. Burke-Spolaor,et al.  The High Time Resolution Universe Pulsar Survey - VI. An artificial neural network and timing of 75 pulsars , 2012, 1209.0793.

[42]  L. Saripalli.  UNDERSTANDING THE FANAROFF–RILEY RADIO GALAXY CLASSIFICATION , 2012, 1206.6893.

[43]  R. Ekers,et al.  The local radio-galaxy population at 20 GHz , 2013, 1304.0268.

[44]  P. Best,et al.  The relation between morphology, accretion modes and environmental factors in local radio AGN , 2013, 1301.1526.

[45]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[46]  C. Flynn,et al.  SPINN: a straightforward machine learning solution to the pulsar candidate selection problem , 2014, 1406.3627.

[47]  Joan Bruna,et al.  Training Convolutional Networks with Noisy Labels , 2014, ICLR 2014.

[48]  Neil Davey,et al.  Teaching a machine to see: unsupervised image segmentation and categorisation using growing neural gas and hierarchical clustering , 2015, 1507.01589.

[49]  Jeff Wagg,et al.  Advancing Astrophysics with the Square Kilometre Array , 2015 .

[50]  Israel,et al.  The new class of FR 0 radio galaxies , 2015, 1510.04272.

[51]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[52]  H. Andernach,et al.  Radio Galaxy Zoo : host galaxies and radio morphologies derived from visual inspection , 2015, 1507.07272.

[53]  H. Falcke,et al.  Nature and evolution of powerful radio galaxies at z ∼ 1 and their link with the quasar luminosity function , 2014, 1411.2968.

[54]  Sander Dieleman,et al.  Rotation-invariant convolutional neural networks for galaxy morphology prediction , 2015, ArXiv.

[55]  H. Andernach,et al.  Radio Galaxy Zoo: discovery of a poor cluster through a giant wide-angle tail radio galaxy , 2016, 1606.05016.

[56]  Ben Hoyle,et al.  Measuring photometric redshifts using galaxy images and Deep Neural Networks , 2015, Astron. Comput..

[57]  David R. Thompson,et al.  A Machine Learning Classifier for Fast Radio Burst Detection at the VLBA , 2016, 1606.08605.

[58]  P. Padovani.  The faint radio sky: radio astronomy becomes mainstream , 2016, The Astronomy and Astrophysics Review.

[59]  F. Massaro,et al.  FRICAT: A FIRST catalog of FRI radio galaxies , 2016, 1610.09376.

[60]  Massimo Brescia,et al.  Cooperative photometric redshift estimation , 2016, Astroinformatics.

[61]  Edward J. Kim,et al.  Star-galaxy Classification Using Deep Convolutional Neural Networks , 2016, ArXiv.

[62]  Y. Wadadekar,et al.  From Nearby Low Luminosity AGN to High Redshift Radio Galaxies: Science Interests with Square Kilometre Array , 2016, 1610.08174.

[63]  Massimo Brescia,et al.  METAPHOR: a machine-learning-based method for the probability density estimation of photometric redshifts , 2016, 1611.02162.