Cannot hand over our FR5 faceplate examples with these guessed scratch labels

Both match the same retained plate; quality has its ID, and I have kept the full images rather than only the marked crops

That helps. Who made the labels? They should see the physical comparison too, otherwise the next image folder may acquire the same problem.

Ask tooling to check the presentation alongside the surface assessment. Moving reflections suggest a capture question, but they do not settle whether a real mark is also present.

Felix, our technician labelled the images and is joining quality; Nathan, tooling is comparing the seating because the full views show one corner in different positions

Keep a confirmed faint defect in that comparison too. An accepted shiny plate alone could reward a setting that simply hides everything troublesome.

Has quality decided the retained plate's condition now, Louis?

Accepted under its visual method, Felix; a separate faint reject is retained for the comparison, and the unsupported scratch labels have been removed from the working set with their history kept

Record the agreed viewing method and storage owner with those references. The next maintainer should be able to find the same identified parts and understand their decisions without reconstructing the meeting.