Further Studies into Synthetic Image Generation using CameoSim

摘要

Previous work by DSTO described the synthetic image generation process using CameoSim and RadThermIR; this report describes the progress made on the modelling of atmospheres, terrains and sensors. A major source of error in the synthetic image generation process previously identified has been resolved. The error concerns the use of interpolation in CameoSim when it uses MODTRAN to incorporate atmospheric effects in synthetic images. A software tool has been created which iteratively reduces the interpolation errors until they are insignificant. Validation studies are currently being planned in the visible to shortwave infrared. In preparation of the validation effort a study of BRDF models has been completed, which includes the physical plausibility of models, how measured data is fitted to models and how well CameoSim samples each model.