I've kept the questioned streak labels separate from confirmed defects in our brushed-trim-plate set. They move with presentation, but that doesn't settle the surface. What should quality record on the retained plates so imaging has something defensible to compare?
Identify each plate, its orientation and the exact accepted or rejected region under the agreed appearance method. Preserve uncertain decisions as uncertain. Then compare those same regions under recorded lighting and actual seating variation, including accepted bright surfaces and defects that become hard to see.
13分