Dortmund 2021 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

AKPIK: Arbeitskreis Physik, moderne Informationstechnologie und Künstliche Intelligenz

AKPIK 2: AKPIK II: Deep Learning

AKPIK 2.9: Talk

Wednesday, March 17, 2021, 18:00–18:15, AKPIKa

Evaluation of Interferometric Data Reconstructed by Neural Networks — •Felix Geyer and Kevin Schmidt — TU Dortmund, Dortmund, Deutschland

Radio interferometry is used to monitor and observe distant astronomical sources and objects with high resolution. Especially Very Long Baseline Interferometry allows achieving the highest resolutions by combining the data of multiple telescopes. This results in an effective diameter corresponding to the greatest distance between two telescopes. The taken data consists of visibilities, which depend on the baselines between the telescopes. Because the distribution of these baselines is sparse, the sample of visibilities is incomplete. After transforming this sample to spatial space, this so-called "dirty image" is inadequate for physical inference and analyses. Thus, the image undergoes an elongated and mostly manually performed cleaning process in order to remove background artifacts and to restore the original source distribution.

A new and fast approach to reconstruct missing data reasonably is using Neural Networks. In this talk, the results and evaluation methods obtained using the simulations created in the radionets framework are presented and discussed. Especially the performance of clean versus noisy input data is examined.