NT & RP Journal
NOVEL PRECISION ENHANCEMENT ALGORITHM WITH REDUCED IMAGE NOISE IN COSMIC MUON TOMOGRAPHY APPLICATIONS
......: info......: history......: editorial......: archive......: for authors......: subscription
Image NTRP

Vol. XXXI, No. 1, Pp. 1-109
March 2016
UDC 621.039+614.876:504.06
ISSN 1451-3994

Back to Contents

 


Pages: 51-64

Authors:
Sangkyu Lee, Amanda Foley, and Tatjana Jevremovic

Abstract


In this paper, we present a new algorithm that improves muon-based generated tomography images with increased precision and reduced image noise applicable to the detection of nuclear materials. Cosmic muon tomography is an interrogation-based imaging technique that, over the last decade, has been frequently employed for the detection of high-Z materials. This technique exploits a magnitude of cosmic muon scattering angles in order to construct an image. The scattering angles of the muons striking the geometry of interest are non-uniform, as cosmic muons vary in energy. The randomness of the scattering angles leads to significant noise in the muon tomography image. GEANT4 is used to numerically create data on the momenta and positions of scattered muons in a predefined geometry that includes high-Z materials. The numerically generated information is then processed with the point of closest approach reconstruction method to construct a muon tomography image; statistical filters are then developed to refine the point of closest approach reconstructed images. The filtered images exhibit reduced noise and enhanced precision when attempting to identify the presence of high-Z materials. The average precision from the point of closest approach reconstruction method is 13 %; for the integrated method, 88 %. The filtered image, therefore, results in a seven-fold improvement in precision compared to the point of closest approach reconstructed image.

Key words: muon tomography, multiple Coulomb scattering, GEANT4, nuclear material

FULL PAPER IN PDF FORMAT (4.69 MB)

Vinča Institute of Nuclear Sciences :: Designed by milas :: July 2007
Operated by acapanic :: Last updated on April, 2016