Nuclear Security

National Security implementation stands on two pillars: safeguards of existing nuclear material and portal monitors for detection of potentially smuggled sources at borders.

We are developing advanced algorithms based on the combination of Bayesian analysis and unsupervised convolutional neural networks.

These algorithms will enable:

  • the identification of neutron sources in the presence of intense gamma-ray background and random or systematic noise sources
  • the use of spectroscopic portal monitors based on cost-effective but low-resolution plastic scintillators
Collaborators: Prof. Yoann Altmann, Heriot-Watt University, Edinburgh, UK, Prof. Al Hero, Electrical Engineering and Computer Science and Prof. Sara Pozzi, Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, MI.