| Summary: | This dissertation was a two-phase effort. Phase I investigated the physical image quality of 16 mesh, etched, and quarterwave antireflection filters for varying levels of filter transmissivity. Three levels of ambient lighting and two levels of monochrome CRT resolution were combined factorially with the filters. In addition, user measures of readability, legibility, and perceived image quality were obtained for these same filter and environmental conditions. Quantitative models were developed to predict the performance and subjective data based on signal and noise measures derived from the physical measurements. Phase II examined the effects of a wide range of filter transmissions and diffuse illuminance on measured image quality and the same user measures as in Phase I.
Phase I showed that while none of the glare filters yielded improved readability or legibility over a baseline condition, the etched and low transmission filters were notable for their degradation of human performance. Mesh and quarterwave filters were found to improve perceived image quality when a specular glare source was present. Modeling was minimally successful for the reading and legibility tasks, but yielded good fit models for perceived image quality. Phase II showed that when even extreme losses in display contrast occurred, users were capable of good reading and legibility performance. Perceived image quality was inversely related to illuminance level. Prediction of performance by image quality metrics was generally not too successful.
It was concluded that in office-type environments, mesh or quarterwave filters can be used to improve perceived image quality when specular glare sources are present, but that no anti-glare filters yielded enhanced short-term readability or legibility over a baseline. Etched filters were not recommended. Measures of physical image quality proved to be good predictors of perceived image quality, but not of timed measures of readability or legibility. Under moderate lighting conditions, monochrome CRTs should be fitted with fairly high transmission filters as it was found the contrast enhancement offered by low transmission filters had negligible effects on performance. Finally, consistent and repeatable findings of degraded legibility for high luminance contrast levels (low illuminance) generated questions as to the existing standards regarding maximum contrast requirements for CRT use. === Ph. D.
|
|---|
