The Effect of Neutron Imaging Pinholes on Image Blurring

TL;DR

This study investigates how realistic pinhole geometries in neutron imaging minimally affect image size, with simulations showing only slight increases in image blurring and suggesting simple models are sufficient for most cases.

Contribution

The paper demonstrates that realistic pinhole geometries have a negligible impact on image blurring in neutron imaging, validating simplified models for practical use.

Findings

Pinhole blurring causes up to 1.5% increase in image area.

Difference between perfect and tapered pinholes is less than 1%.

Simple ray-tracing approximations match full MCNP simulations closely.

Abstract

We examine the effect of realistic pinhole geometries used in neutron imaging on the size of the image produced in the image plane. Using a simple inverted parabola shape for the neutron source, an MCNP simulation shows that the area of the images are only very slightly changed by pinhole blurring. Using a metric, A80, which is the area of the image that contains 80% of the neutrons, we find a maximum of 1.5% increase in the metric. In comparing a `perfect' cylinder bore hole approximation for the pinhole with a full stepped collimator tapered pinhole we find a difference in A80 of 0.8%. In addition, we find very little difference in the prediction of a full MCNP simulation compared to a simple ray-tracing calculation that includes neutron absorption in the pinhole system but does not include neutron scattering. We note that more sophisticated assumptions for the neutron source shape than made here could lead to larger increases in the image size, and increases in A80 of the order of 5-8% could be realistic.