Imaging simulation as a tool for imager performance assessment
Fraunhofer IOSB develops and employs different types of imaging simulation software for studying the performance of thermal imagers and active electro-optical systems. These tools generate images that a defined imaging system would output, based on an undistorted high-resolution input image of the scene. Therefore, the input image is modified step by step according to the effects of each component of the imaging chain. Depending on the considered spectral band, the pixel values of the input image may correspond to the temperatures of the object in the scene or have to be calculated from the reflectivity properties of the objects and the illumination sources.
For the performance assessment of electro-optical systems, imaging simulations offer advantages and disadvantages compared to analytical models such as TRM4. The biggest benefits come from the usage of image data that allows for a realistic simulation of anisoplanatic and statistical effects such as turbulence as well as the simulation of any advanced digital signal processing algorithm integrated in the system. However, the assessment of electro-optical system performance requires to evaluate the quality of the resulting images, which is an art in itself. While image quality metrics can be used to rank the quality of the resulting images and thus the imager performances, the calculation of the range performance is even more complex as it requires a realistic model of the human perception process for the considered scenario.