Our example bezel images have bright streaks labelled scratches, though slight presentation changes move them. I want the physical reference decision settled before threshold work, with repeatable images to match. What should I ask quality to record about the actual surface?
Two are retained with identifiers. Quality has agreed to inspect both against the appearance requirement. The image labels were guesses from our planning session, not their decisions.
Keep the accepted reference as well as any reject. The imaging comparison must reveal the required defect without turning normal reflective features into failures; an impressive reject image alone would not establish that.
Quality accepted one and rejected the other for an edge nick. I have their record and the marked location. Neither broad streak was identified as the reason for rejection.
Include the holder and lighting conditions in that comparison. A cheap first check is to capture both identified pieces repeatedly with the settings recorded, then reseat them and compare what changes. Preserve the original files for review.
Anika, not consistently in our existing images. Noah, we have arranged that repeat comparison with quality and the fixture person. Thank you for separating it from buying a different camera; we have not justified that purchase yet.
Record which part variation that small comparison covers. It can resolve an initial imaging question without yet establishing performance over every bezel you intend to inspect.