Neutron Image Reconstruction Algorithms: Parameter Optimization

Overview

Radiography is a non-destructive technique used to visualize the interior of objects using various types of radiation. Radiography is used in fields such as biology, chemistry, engineering, material science, and archeology for samples including fuel cells, engine parts, artifacts, and plants. Neutron imaging utilizes a beam of neutrons to visualize the interior of high contrast, light and fluid materials. Images obtained by neutron imaging have greater contrast as compared to computed tomography (CT) images, but it is a time and cost- intensive process. The raw data obtained from neutron imaging must be reconstructed using algorithms such as the Filtered Back-Projection (FBP) and Model Based Iterative Reconstruction (MBIR). FBP is an analytical reconstruction algorithm that requires images obtained at many angles. MBIR is more time-consuming than FBP, but the resulting reconstructed image is more precise when fewer angles are available. Using samples of bone scaffolding and a meteor, this work optimizes parameter values (number of angles and signal to noise ratio) in each of the reconstruction algorithms. Reconstructing the meteor to a high-quality image using FBP was optimized at 512, the maximum available. However, only 37 angles of the meteor were needed in MBIR to produce a high-quality reconstruction. The optimal snr parameter value for the scaffold using MBIR was determined to be 25 while for the meteor it was 45. These results can be attributed to the sample’s elemental makeups.

Competition history

• JSHS 2022 Category not listed

Resources

• National JSHS abstract book ↗

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

• JSHS None projects
• Every JSHS 2022 project