Repository of University of Nova Gorica

Show document
A+ | A- | Help | SLO | ENG

Title:Localisation and classification of gamma ray sources using neural networks
Authors:ID Oetelaar, Chris van den (Author)
ID Bhattacharyya, Saptashwa (Author)
ID Panes, Boris (Author)
ID Caron, Sascha (Author)
ID Zaharijas, Gabrijela (Author)
ID Ruiz de Austri, Roberto (Author)
ID Jóhannesson, Guõlaugur (Author)
.pdf ICRC2021_663.pdf (997,42 KB)
MD5: 1AF7F3E79DF10697558A73D85F38CCA6
Work type:Unknown
Typology:1.08 - Published Scientific Conference Contribution
Organization:UNG - University of Nova Gorica
Abstract:With limited statistics and spatial resolution of current detectors, accurately localising and separating gamma-ray point sources from the dominating interstellar emission in the GeV energy range is challenging. Motivated by the challenges of the traditional methods used for the gamma-ray source detection, here we demonstrate the application of deep learning based algorithms to automatically detect and classify point sources, which can be applied directly to the binned Fermi-LAT data and potentially be generalised to other wavelengths. For the point source detection task, we use popular deep neural network structure U-NET, together with image segmentation, for precise localisation of sources, various clustering algorithms were tested on the segmented images. The training samples are based on the source properties of AGNs and PSRs from the latest Fermi-LAT source catalog, in addition to the background interstellar emission. Finally, we have created a more complex but robust training data generation exploiting full detector potential, increasing spatial resolution at the highest energies.
Keywords:gamma-rays, deep learning, computer vision
Year of publishing:2021
Number of pages:str. 1-9
PID:20.500.12556/RUNG-6833 New window
COBISS.SI-ID:78766339 New window
ISSN on article:1824-8039
Publication date in RUNG:01.10.2021
Copy citation
Average score:(0 votes)
Your score:Voting is allowed only for logged in users.
Share:Bookmark and Share

Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Record is a part of a proceedings

Title:37th International Cosmic Ray Conference [also] ICRC2021
COBISS.SI-ID:69435907 New window

Record is a part of a journal

Title:Proceedings of science
Shortened title:Pos proc. sci.
COBISS.SI-ID:20239655 New window


License:CC BY-NC-ND 4.0, Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
Description:The most restrictive Creative Commons license. This only allows people to download and share the work for no commercial gain and for no other purposes.