English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

In this work we explore the possibility of applying machine learning methodsdesigned for one-dimensional problems to the task of galaxy imageclassification. The algorithms used for image classification typically rely onmultiple costly steps, such as the Point Spread Function (PSF) deconvolutionand the training and application of complex Convolutional Neural Networks (CNN)of thousands or even millions of parameters. In our approach, we extractfeatures from the galaxy images by analysing the elliptical isophotes in theirlight distribution and collect the information in a sequence. The sequencesobtained with this method present definite features allowing a directdistinction between galaxy types, as opposed to smooth S\'ersic profiles. Then,we train and classify the sequences with machine learning algorithms, designedthrough the platform Modulos AutoML, and study how they optimize theclassification task. As a demonstration of this method, we use the secondpublic release of the Dark Energy Survey (DES DR2). We show that by applying itto this sample we are able to successfully distinguish between early-type andlate-type galaxies, for images with signal-to-noise ratio greater then 300.This yields an accuracy of $86\%$ for the early-type galaxies and $93\%$ forthe late-type galaxies, which is on par with most contemporary automated imageclassification approaches. Our novel method allows for galaxy images to beaccurately classified and is faster than other approaches. Data dimensionalityreduction also implies a significant lowering in computational cost. In theperspective of future data sets obtained with e.g. Euclid and the Vera RubinObservatory (VRO), this work represents a path towards using a well-tested andwidely used platform from industry in efficiently tackling galaxyclassification problems at the peta-byte scale.

Image feature extraction and galaxy classification: a novel and efficient approach with automated machine learning