An interactive simulation environment for end-to-end digital imaging system design and fidelity analysis
The detailed specification, implementation, and documentation of an interactive software environment based on a continuous/discrete/continuous imaging system model is presented. The purpose of the interactive environment is to support the design and performance analysis of end-to-end digital imaging...
| Main Author: | |
|---|---|
| Format: | Others |
| Language: | English |
| Published: |
W&M ScholarWorks
2000
|
| Subjects: | |
| Online Access: | https://scholarworks.wm.edu/etd/1539623980 https://scholarworks.wm.edu/cgi/viewcontent.cgi?article=3626&context=etd |
| Summary: | The detailed specification, implementation, and documentation of an interactive software environment based on a continuous/discrete/continuous imaging system model is presented. The purpose of the interactive environment is to support the design and performance analysis of end-to-end digital imaging systems. Development of the environment is based on the objectives of acceptable response time, large sampling grid capability, good graphical user interface design, independence from proprietary applications and portability among UNIX workstations. While one-dimensional variations of interactive design environments have been developed by the commercial active filter design community, there is little or no evidence that the increased complexity associated with the extension to two dimensions had been satisfactorily accomplished prior to the work in this dissertation. The computer time versus computer memory trade-off is discussed as it applies in this particular context, and the results of a systematic study of representation passband limits are presented. The object of the study was to determine the representation passband parameters beyond which any aliasing contribution from frequencies beyond the representation passband is invariably negligible. Validation of the environment is documented by an exhaustive consideration of simple input scenes comprised of a uniform square on a uniform background, in which the square can be arbitrarily small and arbitrarily located within the scene. The effects of sampling and the dependence of those effects on sample-scene phase are illustrated in 1-D, used as a predictor for the 2-D outcome, and then illustrated in 2-D for the purpose of comparing the projected and actual results. |
|---|