Stochastic Imaging Algorithms: Stochastic Origins Ensembles (SOE)

Research Focus Areas: 6

Ph.D. Students: M.C. Hamel


To maximize the performance of the dual-particle imager, advanced imaging algorithms are required. These algorithms use statistical methods to remove sources of uncertainty and system resolution to produce a more accurate representation of the true source distribution. A stochastic reconstruction process known as stochastic origin ensembles (SOE) is being investigated as a method to address some of the shortfalls with the popular maximum-likelihood reconstruction methods.

SOE has been used to reconstruct images of multiple radiation sources including special nuclear material. The algorithm has been deigned to produce localized energy spectra for both neutrons and gamma-rays, which allows for source discrimination. Ongoing work includes the optimization of algorithm parameters and characterization of performance in realistic non-proliferation applications.

Stochastic Origins Ensembles with labels of SNM, WGPu, 252 Cf, AmBe, Non-weapons usable material and Thick Lead Shielding

Related Journal Articles:

  1. M.C. Hamel, J. K. Polack, A Poitrasson-Rivière, S. D. Clarke, S. A. Pozzi, “Localization and spectral isolation of special nuclear material using stochastic image reconstruction,” Nuclear Instruments and Methods in Physics Research Section A, vol. 841, pp. 24-33, 2017.
  2. M.C. Hamel, J. K. Polack, A. Poitrasson-Rivière, M. Flaska, S. D. Clarke, S. A. Pozzi, “Stochastic image reconstruction for a dual-particle imaging system,” Nuclear Instruments and Methods in Physics Research Section A, vol. 810, pp. 120-131, 2016.

Related Conference Proceedings:

  1. M.C. Hamel, J. K. Polack, A. Poitrasson-Rivière, S. D. Clarke, S. A. Pozzi, “Stochastic image reconstruction for non-proliferation applications,” IEEE Nuclear Science Symposium Conference Record, San Diego, CA, USA. 31 Oct. – 7 Nov., 2015.