You don't go to the quality show I'd guess.
If you have not seen this you are a good 30+ years behind the curve.
Z axis misalignment or rotation of the coordinate systems requires knowing the Z.
This can be done by a laser spot, line, cross or projected grid or even just an auto focus on a few edges. Obviously back lighting does not work here.
It is no different than the rotation in X-Y.
Software and ease of use is much advanced since my early days in the 80s with the View RB1 and 1220.
A severe limitation vs real optical comparators in the past has simply been display screen resolution and camera pixel counts which is still not quite there yet.
This guy scans the part at a high resolution, it's ability to not see dirt or quality of edge I have no idea.
Most automated system see dirt, dust, minor burrs. A real human throws these away when measuring. Good software tries to do this also but not close to the ability of a human.
Dust, dirt flakes, I've spent a lot of time on removal for the measured true output. A human can measure a dirty part with ease and spot it, not so much here in the computer.
On my "home builds" I program a moving overlay that tracks the slides so you can move to different spots on the part and still see what a traditional overlay would show you.
Often simply called video cmms.
As display and camera pixels counts grow these will become more common.
I gut old comparators and put in a camera, LCD screen and the machine vision software but still leave some untouched and sometimes or many times that old school is better to use.
What I do love is to go the the shows, watch a demo run, ask to make a minor change and then say "let me see you do this".
Yes I am officially a jerk but other people made me look like an ass many times in this field.
Make no doubt that to the old guy doing this before there where desktop computers the posted system is very impressive, very easy to use and just so plain slick.
Bob
