How do we tune an FR10 inspection around a cover nobody kept?

CallumChan1056 · 10 Oct 2025, 09:08 UTC

Closed
CA
CallumChan1056
The FR10 trial folder calls a bright streak a scratch, although it moves between views of the same polished cover. The cover was returned to the production pile. I have asked people to stop using those labels until we have physical references we can identify and keep.

11 replies

LE
LeoChen1145
Replying to CallumChan1056

Leave those images as unverified examples. Build the comparison from identified accepted and rejected covers inspected under the agreed appearance method, with the imaging setup recorded.

12 points
OS
OscarAbbott0010
Replying to CallumChan1056

Do you still have the original image files? Our supposed scratch changed character when we opened the source instead of the presentation copy. Still needed the part, but it explained why people were arguing.

1 points
HE
HenryAli0177
Replying to OscarAbbott0010

Oscar, that can explain a display difference, but without the original cover it cannot resolve its physical condition. Callum should not retrospectively classify it from whichever image now looks more convincing.

14 points
OS
OscarAbbott0010
Replying to HenryAli0177

Yes, that's what I meant by still needing the part. Keep it unknown. I was asking whether the file history could explain how the label spread.

1 points
CA
CallumChan1056
Replying to OscarAbbott0010

Oscar, source files exist. They include the different presentations, not just crops. Henry, we are leaving that cover unclassified. Quality has retained new examples but disagrees about one borderline finish.

23 points
LE
LeoChen1145
Replying to CallumChan1056

Keep the borderline finish in an undecided group until quality settles it. Don't force a binary label just because the software folder wants two directories.

13 points
IM
ImranAdams0131
Replying to CallumChan1056

Our training examples once omitted the difficult finish entirely. The test looked excellent. Include it once its judgement is agreed, not only immaculate parts and obvious damage.

17 points
HE
HenryAli0177
Replying to ImranAdams0131

Imran, agreed, with the intended range of surface and presentation variation represented. The sample selection needs to match what production will actually send to the station.

16 points
OS
OscarAbbott0010
Replying to CallumChan1056

Callum, can the inspector and vision person look at the borderline cover together? Might be cheaper than another week of passing photographs around and each imagining a different mark.

3 points
CA
CallumChan1056
Replying to OscarAbbott0010

They did. The disputed line is acceptable under our finish method; a separate shallow scratch is not. Both retained. Imaging comparison is still open because the scratch is weak in the proposed view.

15 points
LE
LeoChen1145
Replying to CallumChan1056

Keep that weak scratch as a required comparison case. A new view needs to reveal it, not merely make the accepted cover prettier.

15 points

Discussion closed

This discussion is closed to new replies after six months without activity. Last activity: .