The marked scratch changes position between our cover images

LinBell0618 · 25 Mar 2026, 17:05 UTC

Reply to discussion
LI
LinBell0618
We have retained accepted and rejected polished covers for the proposed UR5e inspection station. Quality has identified the rejected feature on the physical cover. That part is settled. The image folder is less convincing. The arrow beside that feature moves between captures, following a bright band rather than the same place on the cover. Our presentation notes also say only seated. I want to rebuild this small reference set before anyone adjusts a threshold around the arrows. What should stay fixed, and what variation should we deliberately include in the comparison?

14 replies

LE
LeoBell0623
Replying to LinBell0618

Check the seat first. Our spare support once hid the corner we needed to inspect.

12 points
LI
LinBell0618
Replying to LeoBell0623

The seat exposes the feature. The two captures use the same support, but the cover can rock slightly before it is fully located.

21 points
BE
BenBrown0892
Replying to LinBell0618

Then seated needs a better definition. Can the person loading it recognise a properly located cover without looking at the camera image for reassurance?

18 points
FE
FelixBaker0458
Replying to LinBell0618

Don't make a perfect seat the whole answer. An accepted polished cover still has reflections. Include that one when the vision person compares lighting, otherwise the reference set teaches an easy problem.

-9 points
LE
LeoBell0623
Replying to FelixBaker0458

Agreed. Locate it consistently, then find out which reflections remain.

11 points
LI
LinBell0618
Replying to BenBrown0892

We checked loading with the fixture designer. One locating contact was obscured in the instruction photo. A new view shows it clearly, and both loaders can now check the cover is seated. The bright band still moves across the accepted cover with small presentation changes.

7 points
LU
LuisArcher0434
Replying to LinBell0618

Who has to make those small changes in the eventual inspection sequence, or are they only part of this comparison?

21 points
BE
BenBrown0892
Replying to FelixBaker0458

Felix, I wasn't claiming the seat would cure reflections. I want the loaders to reproduce a comparison somebody can interpret.

20 points
FE
FelixBaker0458
Replying to BenBrown0892

Fair enough. Lin, has the vision person seen the originals without arrows? I'd hate to find the entire conversation is about somebody's annotation rather than the captured feature.

13 points
LI
LinBell0618
Replying to LuisArcher0434

Luis, these changes are for comparing the imaging options, not an agreed operator sequence. Felix, yes, the originals are retained. Quality has now corrected the arrow against the physical feature; the old arrow was on the adjacent reflection.

11 points
LU
LuisArcher0434
Replying to LinBell0618

Then label the comparison poses as comparison poses; our teaching notes once turned a trial variation into an extra loading instruction.

15 points
LI
LinBell0618
Replying to LuisArcher0434

Done. The revised folder has the identified covers, original captures and the corrected feature mark. The accepted cover with the moving band remains in it. We are comparing two lighting arrangements; neither has been selected for the cell yet.

5 points
BE
BenBrown0892
Replying to LinBell0618

Thanks for coming back with the actual arrow error. Did the old marked images get removed from the teaching handout too?

25 points
LI
LinBell0618
Replying to BenBrown0892

Yes. The tutor opened the revised handout with us and matched its examples to the corrected folder. The imaging choice remains open, but the lesson no longer calls that band a scratch.

15 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.