Hubert Bretonnière | Searcharxiv

arXiv2026

Photometric Super-Resolution for Improving Galaxy Morphological Measurements using Conditional Generative Adversarial Networks

The measurement of galaxy morphological parameters from astronomical images features in a wide range of modern analyses, including galaxy evolution and cosmological weak lensing studies. The precision and accuracy of morphological parameter estimation can be influenced by several key factors. The effective seeing of the image, summarized by the point spread function (PSF), limits how galaxy features or light profiles are resolved. The pixel scale of the detector also influences the resolution and the amount of statistical information available for a given object. The depth of the observations determines the signal-to-noise ratio of the image. Improving each of these factors is very costly, either in terms of detector upgrades, observatory design, or observing time. Here, we develop a conditional generative adversarial network, called Neo, trained to transform existing ground-based images into sharper, finer-scale images comparable to space-based image quality. We demonstrate that Neo improves the accuracy of measured morphological parameters by factors of $2$-$10$ when trained to translate Subaru Hyper Suprime-Camera (HSC) images to approximate Hubble Space Telescope (HST) data. Neo is designed for applicability to ongoing, large-scale surveys such as the Legacy Survey of Space and Time (LSST) conducted by Vera C. Rubin Observatory in combination with space telescopes such as HST, James Webb Space Telescope, and Nancy Grace Roman Space Telescope. These results suggest that Neo could be used to improve both cosmological and galaxy evolution analyses based on massive, ground-based survey datasets like LSST. The model code is open source and available at https://purl.archive.org/neo/code.

astro-ph.IM↗

arXiv2021

Probabilistic segmentation of overlapping galaxies for large cosmological surveys

Encoder-Decoder networks such as U-Nets have been applied successfully in a wide range of computer vision tasks, especially for image segmentation of different flavours across different fields. Nevertheless, most applications lack of a satisfying quantification of the uncertainty of the prediction. Yet, a well calibrated segmentation uncertainty can be a key element for scientific applications such as precision cosmology. In this on-going work, we explore the use of the probabilistic version of the U-Net, recently proposed by Kohl et al (2018), and adapt it to automate the segmentation of galaxies for large photometric surveys. We focus especially on the probabilistic segmentation of overlapping galaxies, also known as blending. We show that, even when training with a single ground truth per input sample, the model manages to properly capture a pixel-wise uncertainty on the segmentation map. Such uncertainty can then be propagated further down the analysis of the galaxy properties. To our knowledge, this is the first time such an experiment is applied for galaxy deblending in astrophysics.

astro-ph.IM↗