Date of Award

3-10-2010

Document Type

Thesis

Degree Name

Master of Science

Department

Department of Engineering Physics

First Advisor

Michael R. Hawks, , PhD.

Abstract

A field deployable hyperspectral imager utilizing chromotomography (CT), with a direct vision prism (DVP) as the dispersive element, has been constructed at AFIT. This research is focused on the development and demonstration of the CT imager. An overview of hyperspectral imaging, chromotomography, a synopsis of reconstruction algorithms, and other CT instruments are given. The importance of component alignment, instrument calibration, and exact prism angular position data are discussed. A simplistic \shift and add" reconstruction algorithm was utilized for this research. Although limited in its ability to reconstruct a spatially and spectrally diverse scene, the algorithm was adequate for the testing and characterization of the CT imager. The AFIT instrument is currently the fastest known DVP based hyperspectral CT imager and is a prototype for a planned space-based system. The instrument has the ability to capture spatial and spectral data of static and transient scenes. Spectral and spatial reconstructions of static scenes are presented in the Experimental Results and Analysis section of this paper. These reconstruction illustrate the effectiveness of the instrument to collect spatial and spectral data. More importantly, the imager can capture spectral data of rapidly evolving scenes such as explosions. The spectrum of a transient event, a firecracker explosion, lasting approximately 0.12 s is presented. Spectral results of the explosion show potassium and sodium emission lines present during the explosion and an absorption feature as the fireball extinguishes. Spatial and spectral reconstruction of a scene in which an explosion occurs during the middle of the collection period is also presented in the Experimental Results and Analysis section of this paper.

AFIT Designator

AFIT-GEO-ENP-10-M01

DTIC Accession Number

ADA516496

Share

COinS