Investigation of One-Sided Radiation-based Approaches to Detection of Visually Obscured Threats

Abstract

In an effort to detect visually obscured objects, a number of different techniques are employed. One such technique is called Coded Aperture Imaging (CAI). Coded apertures operate with no lenses, meaning there is an infinite depth of field, the image does not suffer from chromatic aberration and images can be formed from sources other than visible light, such as high energy radiation. When gamma radiation is projected at an object it will produce a detector response which can be reconstructed to reveal the original object. Computer modeling was performed using a specific binary matrix to generate a three-dimensional mask, using holes for zeros and tungsten for ones. Radiation transport software called MCNPX was used to simulate radiation passing through the holes and registering on the detector. Due to the number of holes, the projection is convoluted, with many copies of the image blurred together. Digital reconstruction techniques were implemented using MATLAB, involving mathematical manipulation in the frequency domain. Since gamma rays are affected by the material density, images consist of shades of grey, resembling medical x-rays. Presently, we can image with transmission x-rays, however, the goal is to image the backscatter from an object which would permit imaging of concealed objects.