I think because they had an accurate a reliable benchmark reference to start with and lots of tweaking before they got it right I bet.
They say "and the high production repeatability for the HC/HCD lenses"
If the image is substantially degraded especially around bright stars there isnt any info to recover anyway - its blown out.
So having good data in the first place would be essential.
But there exists a possibility to do something - one day they will have software and reference images with which the ordinary man will be able to calibrate their scope images against along with th enormal flast and darks etc to improve the odds.
Ever seen Focus Magic - its advanced deconvolution algorithms go some way to this already.
More importantly, how well do these techniques work with pin point light sources such as stars? All algorithms require reference points or lines to judge what corrections need to be made. This is typically fine for terrestrial photography, but considerably complex with stars due to lack of uniformity. How would an algorithm know what's right from wrong? Would be interesting to evaluate.
I am guessing each lens is tested against standard targets in the factory and the settings required to correct for the common problems are probably encoded in a small chip inside the lens body somewhere or in a unique bar code of some sort that the camera reads, and hey presto, the camera software interprets what it gets on the chip and the lens settings to make an approximate compensation that removes most the optical effects.
This costs money, hence the price of Hasselblad stuff.
