Application of statistical pattern recognition and deep learning for morphological classification in radio astronomy

Date
2022-04
Journal Title
Journal ISSN
Volume Title
Publisher
Stellenbosch : Stellenbosch University
Abstract
ENGLISH ABSTRACT: The morphological classification of radio sources is important to gain a full under standing of galaxy evolution processes and their relation with local environmental properties. Furthermore, the complex nature of the problem, its appeal for citi zen 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 to identify whether overfitting occurs when training CNNs to perform radio galaxy morphological clas sification is needed. Secondly, a comparative study regarding the practical appli cability of the CNN architectures in literature is required. Both of these short comings are addressed in this thesis. Multiple performance metrics are used for the latter comparative study, such as inference time, model complexity, compu tational complexity and mean per class accuracy. A ranking system based upon recognition and computational performance is proposed. MCRGNet, ATLAS and ConvXpress (novel classifier) are the architectures that best balance computational requirements with recognition performance.
AFRIKAANSE OPSOMMING: Die morfologiese klassifikasie van radiobronne is belangrik om ’n volledige begrip van die evolusieprosesse binnein sterrestelsels te ontwikkel, asook die rol wat hul plaaslike omgewings hierin speel. As gevolg van die ingewikkelde aard van die probleem, asook die aantrekkingskrag daarvan vir “burgerwetenskaplikes” en die groot hoeveelhede data wat deur bestaande en opkomende radioteleskope gege nereer word, maak die morfologiese klassifikasie van radiobronne ’n ideale proef gebied vir die toepassing van masjienleertegnieke. ’n Benadering wat belowend lyk, is Konvolusionele Neurale Netwerke (KNNe). Literatuur ontbreek egter twee belangrike dinge as dit kom by KNNe en die morfologiese klassifikasie van radio sterrestelsels. Eerstens is daar ’n analise nodig rondom die identifikasie van oor passing wanneer KNNe afgerig word om radio sterrestelsels volgens morfologie te klassifiseer. Tweedens word ’n vergelykende studie oor die praktiese toepaslik heid van die KNN-argitekture in literatuur benodig. Albei hierdie tekortkominge word in hierdie tesis aagespreek. Veelvuldige prestasiemetings word vir laasgenoemde vergelykende studie gebruik, soos inferensietyd, modelkompleksiteit, berekeningkompleksiteit en gemiddelde akkuraatheid per klas. ’n Rangorde skema word voorgestel gebaseer op herkenning en berekeningsprestasie. MCRGNet, AT LAS en ConvXpress (nuwe bydrae) is die argitekture wat berekeningsvereistes en herkenningsprestasie die beste balanseer.
Description
Thesis (MSc)--Stellenbosch University, 2022.
Keywords
Machine Learning, UCTD, Radio astronomy, Pattern recognition systems, Deep learning (Machine learning), Morphological classification
Citation