The scratch labels look less certain than the images suggest

ImranBarnes0566 · 17 Apr 2026, 19:08 UTC

Reply to discussion
IM
ImranBarnes0566
Our planned presenting reflective components for surface inspection in a cosmetic components workshop has Universal Robots UR5e presenting a polished metal cover to a fixed camera. The example images contain bright streaks that shift with small presentation differences, yet those streaks are being labelled as scratches. Before commissioning, I'd like to establish reliable references and image conditions for threshold tuning.

14 replies

AM
AmaraBell0687
Replying to ImranBarnes0566

@ImranBarnes0566 Check whether the suspected scratches were confirmed on physical samples under the agreed appearance procedure. Link each image to its sample identity and inspection result first.

10 points
IM
ImranBarnes0566
Replying to AmaraBell0687

@AmaraBell0687 I've found that some labels were assigned from the images themselves. The samples are still available, but those labels have no corresponding physical inspection results, making the reference circular.

16 points
AM
AmaraBell0687
Replying to ImranBarnes0566

Mark those labels uncertain and have the samples inspected properly. Then compare images with presentation and lighting controlled; the moving streaks are a useful reflection clue.

5 points
MI
MiaChan1079
Replying to AmaraBell0687

That clue needs a limit: presentation can also change how a real scratch appears. Movement in the image isn't enough to reclassify the sample as acceptable.

6 points
AM
AmaraBell0687
Replying to MiaChan1079

Correct. I meant a reason to investigate the imaging, not a new defect label. Physical acceptance stays with the agreed inspection method.

7 points
OW
OwenBennett0747
Replying to AmaraBell0687

So uncertain isn't the same as good? Where do those images go?

21 points
AM
AmaraBell0687
Replying to OwenBennett0747

@OwenBennett0747 Retain their sample identities and undecided status. They shouldn't become confirmed acceptable or defective examples while the physical review remains unresolved.

9 points
IM
ImranBarnes0566
Replying to AmaraBell0687

@AmaraBell0687 I've checked the image folders and found that exposure and presentation both changed between them. Their improved appearance can't be attributed to either change individually.

18 points
WI
WillChan1096
Replying to ImranBarnes0566

@ImranBarnes0566 On my inspection setup, a folder called 'better lighting' also contained easier samples. The folder name was doing a lot of unearned work.

11 points
IM
ImranBarnes0566
Replying to WillChan1096

Which details did you retain with each image? I can record sample identity and settings, but I'd like a practical record people will actually maintain.

6 points
WI
WillChan1096
Replying to ImranBarnes0566

@ImranBarnes0566 On my setup we kept sample ID, exposure, lighting arrangement and presentation condition. Enough to explain a comparison without reconstructing the whole session from memory.

8 points
AM
AmaraBell0687
Replying to WillChan1096

Use those details to ask your vision specialist for controlled comparisons on representative surfaces. Keep the sample set fixed when assessing an acquisition change.

14 points
IM
ImranBarnes0566
Replying to AmaraBell0687

The inspection concept remains undecided because the available references and images don't establish reliable discrimination of the required defects.

11 points
AM
AmaraBell0687
Replying to ImranBarnes0566

@ImranBarnes0566 Your reason for holding the decision is clear: the current evidence doesn't establish reliable identification of the required defects.

4 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.