I have stopped adding reject examples to our FR5 camera folder. The same brushed trim plate appears in three poses, with a bright line in a different place each time. Nobody has identified a corresponding scratch on the part.
We need a defensible reference set before threshold work. I can organise the parts and images; I cannot decide the cosmetic standard on quality's behalf.
Our image review became easier once quality marked the physical feature on an identified sample; can your inspector examine this plate against the acceptance examples before the imaging team changes anything?
Keep that scuff decision separate from the image labels for now, and retain the original captures so the imaging team can explain why the bright lines moved.
Have they settled the scuff decision, Evan? I'd like to know whether you now have an accepted plate with misleading images or an actual reject whose mark needs to remain visible.
Accepted plate. Quality says the edge scuff is within their approved example range. The moving bright lines are not their rejection reason. I've withdrawn these images from the proposed reject set.
Quality has selected accepted and rejected plates with identified features. Camera team is comparing lighting on a fixed presentation first. Repeated loading remains a separate check because the nest review is unfinished.