Authored By Hui Chen
Image Simulation tool is one of the most popular analysis tools in OpticStudio. It provides user a direct visual of the image quality, such as sharpness of the image, distortion, color aberrations, etc. It simulates system imaging performance by convolving a source bitmap file with an array of Point Spread Functions.
In this tool, a source file, which can be a BMP, JPG, IMA, or BIM, is placed in the object space. Field Height in the Image Simulation setting is used to define the full height in the y-axis of the source bitmap in field coordinates, which may be either lens units or degrees. For example, if you use Angle as your Field type in the Field Data Editor, the input value in the Field Height in Image Simulation will also be in angle (degrees). If you want the bitmap source to fill the field of view, you can calculate the Field Height for the Image Simulation as shown in the graph below:
In Image Simulation algorithm, a "grid" of Point Spread Functions (PSFs) is computed. The grid spans the field size and describes the aberrations at selected points in the field of view defined by the bitmap and field size settings. In the Aberrations setting, user can choose whether or not to include diffraction effect. When choosing Diffraction, the algorithm uses the Huygens PSF for convolution with source bitmap. The PSF-X and PSF-Y points control how many Point Spread Functions are used to sample the entire field. The more PSF points you are using, the better the entire field gets sampled, at a cost of longer computation time. The rule of thumb is if your system performance varies significantly as a function of the field, a higher number of PSF points may be required to sufficiently sample the field.
Ideally, you don't want the PSF to change drastically between PSF grid locations. If the shape of the PSF is dramatically different between neighboring grid locations, the 2D interpolation scheme could introduce unphysical artifacts. We recommend checking the PSF Grid, an option in the Show As dropdown, to get a visual on the PSFs used to sample the field. If a PSF grid point appears to only span a single pixel, the Point Spread Function is small compared to the source bitmap pixel size. In this case, the source bitmap could be oversampled by increasing the Oversampling setting. Oversampling increases the pixel resolution of the source bitmap by copying one pixel into 2, 4, or more identical adjacent pixels. This increases the number of pixels per field unit. Or you could try reducing the source bitmap Field Height in order to make the PSF grid large compared to the pixel size of the source bitmap (i.e. it spans multiple pixels). This helps reduce the artifact caused by low pixel resolution of the input source bitmap. That said, it’s recommended the source scene bitmap should have sufficient pixel resolution for your intended application.
Pupil and Image Sampling define the grid sampling to use in the pupil space and the source bitmap space for computing the PSF grid respectively. As for choosing the Pupil sampling, we recommend running the Huygens PSF analysis first to look at the size and shape of the PSF. Because the Image Simulation uses the Huygens PSF, a reasonable sampling used in the Huygens PSF analysis will be a good starting point to use for the Image Simulation Pupil sampling option.
If you would like to know more about the Image Simulation tool, we recommend this KBA written by Mark Nicholson, How to simulate high resolution images. This article provides an interesting, in-depth discussion of how to best use this tool.