Skip to main content

Mission Overview

Classifying galaxy mergers with JWST/HST and CNNs

This entry explores how to classify merging galaxies vs. non-merging galaxies from multi-wavelength imaging from the James Webb Space Telescope (JWST) and the Hubble Space Telescope (HST) with convolutional neural networks (CNNs). This approach was used by the DeepMerge team (Ciprijanovic et al. 2020) to classify mergers in sythetic observations of simulated galaxies. The same workflow is presented in simplifed format to walk through the construction of the CNN model. The results are then validated and the performance is discussed. 

Data: The DeepMerge HLSP

Notebook: Classifying JWST-HST galaxy mergers with CNNs

Released: 2022-06-12

Updated: 2022-06-12

Tags: classification, deep learning, 1d data

Example simulated galaxy images (JWST NIRCam F356W) from DeepMerge (Ciprijanovic et al. 2020).