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

Aims. In astronomy, machine learning has been successful in various taskssuch as source localisation, classification, anomaly detection, andsegmentation. However, feature regression remains an area with room forimprovement. We aim to design a network that can accurately estimate sources'features and their uncertainties from single-band image cutouts, given theapproximated locations of the sources provided by the previously developed codeAutoSourceID-Light (ASID-L) or other external catalogues. This work serves as aproof of concept, showing the potential of machine learning in estimatingastronomical features when trained on meticulously crafted synthetic images andsubsequently applied to real astronomical data. Methods. The algorithmpresented here, AutoSourceID-FeatureExtractor (ASID-FE), uses single-bandcutouts of 32x32 pixels around the localised sources to estimate flux,sub-pixel centre coordinates, and their uncertainties. ASID-FE employs atwo-step mean variance estimation (TS-MVE) approach to first estimate thefeatures and then their uncertainties without the need for additionalinformation, for example the point spread function (PSF). For this proof ofconcept, we generated a synthetic dataset comprising only point sourcesdirectly derived from real images, ensuring a controlled yet authentic testingenvironment. Results.We show that ASID-FE, trained on synthetic images derivedfrom the MeerLICHT telescope, can predict more accurate features with respectto similar codes such as SourceExtractor and that the two-step method canestimate well-calibrated uncertainties that are better behaved compared tosimilar methods that use deep ensembles of simple MVE networks. Finally, weevaluate the model on real images from the MeerLICHT telescope and the ZwickyTransient Facility (ZTF) to test its transfer learning abilities.

astronomy and astrophysics

AutoSourceID-FeatureExtractor. Optical image analysis using a two-step  mean variance estimation network for feature estimation and uncertainty  characterisation