Date of Award
Master of Science
Department of Engineering Physics
Michael R. Hawks, PhD.
A hyperspectral chromotomographic instrument has been developed to determine the spatial and spectral distribution of a target. It does this by recording the dispersed images produced by a rotating direct vision prism with a high speed camera. The recorded images are processed through a tomographic reconstruction algorithm to produce a reconstructed hypercube. A physics-based system model was developed to investigate three areas. The first was how the prism angle selection and range of angles used impacted the reconstructed image quality. The second was an assessment of a new iterative reconstruction algorithm developed to remove artifacts produced in the Back Projection algorithm. The third area was how optical aberrations in the recorded images impacted the reconstructed image. The areas of investigation were completed using a new prism design because a distortion-like effect found in the current prism design. The majority of the image quality gains were accomplished in the first 30 projections used in the reconstruction. Provided at least 30 projections were used in the reconstruction the new algorithm, developed for this investigation, outperformed Back Projection in 15 of the 18 targets evaluated.
DTIC Accession Number
Tervo, Ryan L., "Comparative Analysis of Reconstructed Image Quality in a Simulated Chromotomographic Imager" (2014). Theses and Dissertations. 663.