I cannot approve bezel thresholds from these bright-line labels

CarlaBarnes0575 · 30 Jul 2026, 06:34 UTC

Reply to discussion
CA
CarlaBarnes0575
I'm planning FR5 presentation of aluminium bezels to a fixed camera. Change presentation slightly and the bright streak changes position. Our example set calls those streaks scratches anyway. Before commissioning, how do I establish surface references and imaging conditions worth tuning against?

10 replies

OS
OscarBennett0706
Replying to CarlaBarnes0575

Does the product appearance requirement specify how the physical surface is judged? Put that decision with the responsible inspection person before using any camera label as ground truth.

22 points
CA
CarlaBarnes0575
Replying to OscarBennett0706

It specifies the viewing method. We didn't use that method to label the image set; the meeting just circled bright lines.

21 points
CA
CallumChan1056
Replying to CarlaBarnes0575

That gives you a concrete correction to make. Retain those annotations as provisional, associate them with identified bezels and obtain the physical judgements. It is not yet evidence that every circled line is harmless.

13 points
RE
ReeceAllen0297
Replying to CarlaBarnes0575

In my earlier comparison, inconsistent seating contributed to changing glare. I would inspect repeat presentation here too, once you know which physical features the camera must reveal. Not a diagnosis of your holder.

9 points
OS
OscarBennett0706
Replying to ReeceAllen0297

Reece, can Carla start recording seating conditions while the physical review is pending? Otherwise those images may acquire better labels but still lack capture context.

19 points
RE
ReeceAllen0297
Replying to OscarBennett0706

Yes. I meant the judged references are needed to assess visibility, not that all observation must wait. Keep those two strands moving together.

12 points
CA
CarlaBarnes0575
Replying to ReeceAllen0297

I'll collect the holder and presentation records while quality reviews the retained bezels. Thanks, that avoids booking two separate rounds just to recover the context.

6 points
CA
CallumChan1056
Replying to CarlaBarnes0575

Keep a few independently selected examples outside tuning when you get that far. Otherwise the same small set can become both your target and the evidence that you hit it.

12 points
CA
CarlaBarnes0575
Replying to CarlaBarnes0575

Physical classifications are back. The image labels are corrected, but one rejected feature isn't consistently visible across repeat placements. Threshold work remains on hold.

23 points
OS
OscarBennett0706
Replying to CarlaBarnes0575

Which capture condition changes its visibility? Preserve the rejected reference while reviewing that, rather than replace it with an easier defect to finish the exercise.

20 points

Add to the discussion

Welcome to Application Robot

Everyone can read the forum. Sign in or create an account to start a discussion, reply, or upload photos.

Forgot your password?

By creating an account, you agree to our Terms and Conditions and community guidelines. Read our Privacy Policy for how your information is handled.