Material discrimination using cosmic ray muon scattering tomography with an artificial neural network

Introduction Muon scattering tomography (MST) can be employed to scan cargo containers and vehicles for special nuclear materials by using cosmic muons. However, the flux of cosmic ray muons is relatively low for direct detection. Thus, the detection has to be done in a short timescale with small numbers of muons to satisfy the demands of practical applications.
Method In this paper, we propose an artificial neural network (ANN) algorithm for material discrimination using MST. The muon scattering angles were simulated using Geant4 to formulate the training set, and the muon scatter angles were measured by Micromegas detection system to create the test set.
Results The ANN-based algorithm presented here ensures a discrimination accuracy of 98.0% between aluminum, copper and tungsten in a 5 min measurement of 4 × 4 × 4 cm³ blocks.