Discriminating sub-TeV gamma and hadron-induced showers through their footprints

TL;DR

This paper introduces a novel gamma/hadron discrimination method for sub-TeV gamma-ray observatories using a pre-trained Vision Transformer to analyze shower footprints, showing resilience to noise and adaptability to various conditions.

Contribution

It presents a new approach leveraging Vision Transformers for gamma/hadron discrimination, effective at lower energies where traditional methods struggle.

Findings

Method is resilient to atmospheric muons and low-energy proton showers.

Effective across different zenith angles and array configurations.

Potential for application in current and future gamma-ray observatories.

Abstract

Gamma/hadron discrimination in ground-based gamma-ray observatories at the sub-TeV energy range is challenging as traditional muon-based methods become less effective at lower energies. This work explores a novel gamma/hadron discrimination method for Extensive Air Shower arrays that analyzes the shower signal footprint patterns using a state-of-the-art pre-trained Vision Transformer (ViT). The resilience of the method to background noise, such as atmospheric muons and low-energy proton showers, along with its adaptability to different zenith angles and array configurations, demonstrates its potential for application in current and future ground-based gamma-ray observatories.