A large fleet (in the hundreds) of high quality telescopes are used for tracking and imaging of launch vehicles during ascent from Cape Canaveral Air Force Station and Kennedy Space Center. A maintenance tool has been development for use with these telescopes. The tool requires rankings of telescope condition in terms of the ability to generate useful imagery. It is thus a case of ranking telescope conditions on the basis of the perceptual image quality of their imagery. Perceptual image quality metrics that are well-correlated to observer opinions of image quality have been available for several decades. However, these are quite limited in their applications, not being designed to compare various optical systems. The perceptual correlation of the metrics implies that a constant image quality curve (such as the boundary between two qualitative categories labeled as excellent and good) would have a constant value of the metric. This is not the case if the optical system parameters (such as object distance or aperture diameter) are varied. No published data on such direct variation is available and this dissertation presents an investigation made into the perceptual metric responses as system parameters are varied. This investigation leads to some non-intuitive conclusions. The perceptual metrics are reviewed as well as more common metrics and their inability to perform in the necessary manner for the research of interest. Perceptual test methods are also reviewed, as is the human visual system. iv Image formation theory is presented in a non-traditional form, yielding the surprising result that perceptual image quality is invariant under changes in focal length if the final displayed image remains constant. Experimental results are presented of changes in perceived image quality as aperture diameter is varied. Results are analyzed and shortcomings in the process and metrics are discussed. Using the test results, predictions are made about the form of the metric response to object distance variations, and subsequent testing was conducted to validate the predictions. The utility of the results, limitations of applicability, and the immediate ability to further generalize the results is presented.
| Identifer | oai:union.ndltd.org:ucf.edu/oai:stars.library.ucf.edu:etd-2863 |
| Date | 01 January 2011 |
| Creators | Lentz, Joshua K |
| Publisher | STARS |
| Source Sets | University of Central Florida |
| Language | English |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Electronic Theses and Dissertations |
Page generated in 0.002 seconds