| This dissertation studies rear projection diffusing screens. We provide a methodology, a theoretical background, metrics, and experimental results to aid in the understanding and design of such screens.; For this work, a theoretical model has been developed to predict local fluctuations of measured color that appear as image noise in projection screens. This predictive model is based on Fraunhofer diffraction along with Huygens' wavelet analysis and linear systems theory. Of importance are the figures of merit that have been defined and used to compare the theoretical predictions and experimental results. The range of validity of the model has also been determined.; We set up an experiment to test the theoretical model. By experimentally varying the numerical aperture of the input illumination, color variations on screens have been measured and characterized. The results of both the experiment and the model show a strong relationship between the polychromatic image noise and the size of the illumination cone. As the size of the illumination cone was decreased from 6 to less than 0.1 millisteradians, there was more than a threefold increase in the figures of merit.; Our model shows insight, validates and augments a common rule of thumb. It is often assumed that making the screen structure significantly larger than the coherence length of the source will result in a system with minimal noise. The model shows that this is correct, but it also provides predictions in the cross over region. This allows one to understand how the image noise in a projection system will change as screen designs are changed. Ultimately, this allows screen solutions to be assessed before they are reduced to practice. |
