Regensburg 2013 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
ST: Fachverband Strahlen- und Medizinphysik
ST 1: Poster
ST 1.2: Poster
Mittwoch, 13. März 2013, 11:00–13:00, Poster B2
Minkowski maps: detection of diffuse signals with strong background noise — •Michael A. Klatt1,2, Daniel Göring2, Christian Stegmann3,4, and Klaus Mecke1,2 — 1Institut für Theoretische Physik, Universität Erlangen — 2ECAP, Universität Erlangen — 3DESY — 4Universität Potsdam
A new technique based on the morphology of random fields is developed to detect non-Poisson features via structural deviations; for example for detection of spatially extended signals in count maps with high background noise. The morphometric analysis quantifies the structure of random fields by Minkowski functionals, which are powerful shape measures. Hadwiger’s completeness theorem ensures that Minkowski functionals provide robust and comprehensive shape information and have short computation times. The joint probability distribution of all three Minkowski functionals is determined by a Monte-Carlo-algorithm for a Poisson random field.
Common likelihood ratio methods are exclusively based on the number of excess counts and discard all further structure information. The analysis based on Minkowski functionals incorporates this additional geometric information in an unbiased analysis, i.e. not looking for an explicit shape but detecting any structural deviation [1]. The non-Poisson features, e.g. sources, are visualized with Minkowski maps.
The technique is designed to detect extended sources in γ-ray astronomy and is successfully applied to data of the H.E.S.S. experiment.
[1] D. Göring, M. A. Klatt et al., A&A submitted (2012).