skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Neutron-computer tomography using real-time neutron radiography

Thesis/Dissertation ·
OSTI ID:6858484

A real-time neutron radiography facility was constructed including the capability of neutron tomography. The neutron beam was measured for total neutron flux ((1.0 +/- 0.2) x 10/sup 11/ m/(m/sup 2/-sec)), gold cadmium ratio (52 +/- 3) and effective neutron temperature (83/sup 0/C +/- 8/sup 0/C). The angular divergence or nonparallelism of the neutron beam was measured to be 2.3/sup 0/ +/- 0.1/sup 0/ thereby providing a means of quantifying the collimator effectiveness. The resolution capabilities of both static film and real-time neutron radiographs were quantified using a Fourier transform algorithm to calculate the modulation transfer function of both types of radiographs. The contrast sensitivity of both types of radiographs. The contrast sensitivity of both types of radiographs was measured as 3.1% for film and 4.0% for real-time radiographs. Two tomography algorithms, the simultaneous iterative reconstruction technique (SIRT) and the convolution method, were programmed on an Intellect 100 Image Processing System. The SIRT algorithm was found to be too large and slow on the Intellect 100 to produce useful tomographs. The convolution method produced results near the theoretical resolution limits for a given number of projections. A tomographic resolution of at least 1.3 mm was demonstrated using 200 projections. Computer running time for the convolution method was found to be approx. 30 seconds for each projection used. A series of experiments were conducted using the convolution method investigating the effect of high and low pass filtering, diagonal line enhancement and contrast stretching on the finished tomographs. These experiments showed that spatial resolution is not generally improved by these filtering functions but contrast sensitivity of the tomographs is improved.

Research Organization:
Virginia Univ., Charlottesville (USA)
OSTI ID:
6858484
Resource Relation:
Other Information: Thesis (Ph. D.)
Country of Publication:
United States
Language:
English