Speaker
Description
Muon tomography exploits the natural flux of cosmic-ray muons to probe the internal structure of large or dense objects using scattering and absorption signatures. GEANT4 provides a powerful Monte Carlo framework for modeling the passage of muons and secondary particles through matter, making it an essential tool for designing and optimizing muon imaging systems. Here I present a general approach to developing GEANT4 simulations of muon scattering tomography, outlining the key components of such models: the muon source, the detector geometry and response, the target object, and the relevant physical processes governing muon interactions. Particular attention is given to modeling realistic cosmic-ray muon sources, including angular and energy distributions consistent with measured atmospheric spectra. Several historical and recent simulation studies are discussed to illustrate typical setups and representative results. Finally, I highlight how detailed GEANT4 simulations can provide synthetic training data for machine learning models, facilitating rapid image reconstruction and material identification in muon tomography applications.