Shower-Aware Dual-Stream Voxel Networks for Structural Defect Detection in Cosmic-Ray Muon Tomography

TL;DR

This paper introduces SA-DSVN, a novel 3D convolutional neural network that leverages secondary electromagnetic shower data alongside scattering angles for improved defect detection in concrete via muon tomography.

Contribution

The work demonstrates that incorporating shower multiplicity significantly enhances defect detection accuracy, a novel approach in muon tomography analysis.

Findings

Shower multiplicity alone increases defect-mean Dice from 0.535 to 0.685.

Model achieves 96.3% voxel accuracy and perfect volume detection sensitivity.

Secondary shower data is a highly effective feature for defect detection.

Abstract

We present SA-DSVN, a 3D convolutional architecture for voxel-level segmentation of structural defects in reinforced concrete using cosmic-ray muon tomography. Unlike conventional reconstruction methods (POCA, MLSD) that rely solely on muon scattering angles, our approach jointly processes scattering kinematics (9 channels) and secondary electromagnetic shower multiplicities (40 channels) through independent encoder streams fused via cross-attention. Training data were generated using Vega, a cloud-native Geant4 simulation framework, producing 4.5 million muon events across 900 volumes containing four defect types - honeycombing, shear fracture, corrosion voids, and delamination - embedded within a dense 7x7 rebar cage. A five-variant ablation study demonstrates that the shower multiplicity stream alone accounts for the majority of discriminative power, raising defect-mean Dice from 0.535 (scattering only) to 0.685 (shower only). On 60 independently simulated validation volumes, the model achieves 96.3% voxel accuracy, per-defect Dice scores of 0.59-0.81, and 100% volume-level detection sensitivity at 10 ms inference per volume. These results establish secondary shower multiplicity as a previously unexploited but highly effective feature for learned muon tomographic reconstruction.