Use of generative adversarial neural networks in scattering muography

  • Pablo Martinez Ruiz del Arbol Instituto de F´ısica de Cantabria, Avda. Los Castros s/n, 39005, Santander, Spain
  • Rubén López Ruiz Instituto de F´ısica de Cantabria, Avda. Los Castros s/n, 39005, Santander, Spain
  • Celia Fernández Madrazo University of Boston, Boston, MA 02215, United States
Keywords: Muography, GAN networks, Machine Learning Muography

Abstract

Many muography applications make an extensive use of simulations to determine detectors design or to train imaging or regression algorithms. The computing cost of producing these simulations is usually quite high, specially concerning the interaction of cosmic muons with matter. This work explores the possibility of using Generative Adversarial Neural (GAN) networks to produce a fast and realistic simulation of the multiple scattering process. The results of the network are confronted with GEANT4 simulations using a benchmark problem related to the measurement of the inner wear of industrial pipes. The GAN is able to reproduce the angular distributions and correlations with a speed up factor of roughly 50 with
respect to GEANT4.

Published
2024-04-30
How to Cite
[1]
P. Martinez Ruiz del Arbol, R. López Ruiz, and C. Fernández Madrazo, “Use of generative adversarial neural networks in scattering muography”, Journal of Advanced Instrumentation in Science, vol. 2024, no. 1, Apr. 2024.
Section
International Workshop on Cosmic-Ray Muography (Muography2023), Naples, Italy