Erlangen 2026 – scientific programme

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

T: Fachverband Teilchenphysik

T 106: Gamma Astronomy III

T 106.1: Talk

Friday, March 20, 2026, 09:00–09:15, KS 00.005

Unsupervised Machine Learning for Muon Ring Classification in Imaging Atmospheric Cherenkov Telescopes — •Giovanni Cozzolongo¹, Alison Mitchell¹, and Samuel Spencer² — ¹ECAP, FAU Erlangen-Nürnberg — ²CTAO SDMC, Zeuthen, Germany

Muons from extensive air showers create ring-like patterns in Imaging Atmospheric Cherenkov Telescopes (IACTs). These ring images encode valuable information about muon properties such as energy and direction, making them important for studying the physics of muons in air showers. However, automated identification of muon events remains challenging, particularly for partial rings and mixed events containing both muon rings and air shower components. We present an unsupervised deep learning approach to classify muon events in H.E.S.S. CT5 data using clustering techniques to automatically separate events into categories like complete muon rings, partial rings, mixed events, and standard air showers. This represents one of the first deep learning applications to the analysis of muons in IACT data, offering significant advantages over manual classification and analytical methods that focus primarily on identifying complete rings.

Keywords: muon; IACT; gamma-ray; HESS; deep-learning