In IMAGINE Photogrammetry only one principal point offset can be entered for the camera information. Should it be the principal point of autocollimation (PPA) or principal point of symmetry (PPS)?
IMAGINE Photogrammetry prefers the value of PPS, which is the adjusted point with minimized distortion. This means the value you input is assumed to be the value of PPS.
In terms of when to use PPS or PPA, it is not important if you are going to run triangulation. There is a preference for PPS since errors are supposedly better distributed. In special situations, the following rules should be considered:
If you are going to use an external calibration report with lens distortion parameters, use the principal point the report uses for the lens distortion, most likely PPS.
If you are going to ignore the lens distortion, use PPS since the lens distortion is minimized with PPS.
If you import the triangulation results from an external triangulation package and use the results directly without re-triangulation, use the principal point that package uses.
In IMAGINE Photogrammetry there is only one entry for the principal point. You cannot enter both values and you cannot define which one you will enter (PPS or PPA). This issue appears when we have a camera with both PPS and PPA calibrated separately. Most IMAGINE Photogrammetry users employ camera calibration reports from USGS and the PPA offset values are zero. In this case IMAGINE Photogrammetry prefers that the PPS is used. If the calibration report has non-zero PPA values and no PPS, the PPA values can be used. For a metric camera like those calibrated by USGS, the effect of using PPA and PPS will be marginal because lens distortions are generally small. In practice it normally will not cause any difference, because the small difference in principal point will be converted into a small shift of the projection center through triangulation, so that the ground coordinates will be the same.