简体中文界面。帖子、指南及政策正文保留原文;未对原文作自动翻译。

Fairino FR5: Are we teaching our camera to reject reflections?

LinAllen0270 · 2026年7月25日 16:05 UTC

回复讨论
LI
LinAllen0270
Our planned checking cosmetic metal surfaces with a fixed camera in a precision metal-parts inspection area has Fairino FR5 presenting a brushed stainless trim plate 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.

7 条回复

TO
TobyAllen0318

Were those scratches confirmed on the physical samples using your agreed appearance method? Start by linking each image to a sample and its inspection result.

21
LI
LinAllen0270

Some labels came from the pictures alone. We still have the samples, but those labels aren't tied to physical inspection results. That's irritatingly circular.

2
TO
TobyAllen0318

Keep the image-only labels uncertain pending physical inspection. A controlled comparison of lighting and presentation can then investigate whether the moving streaks indicate reflection effects.

22
LO
LouisChan1094

@TobyAllen0318 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.

1
TO
TobyAllen0318

@LouisChan1094 You're right to qualify that. The streak behaviour motivates an imaging investigation; it doesn't replace physical classification under the agreed inspection procedure.

6
LI
LinAllen0270

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

8
TO
TobyAllen0318

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

17

参与讨论

欢迎来到 Application Robot

所有人都可以阅读论坛。登录或注册后即可发起讨论、回复或上传照片。

忘记密码?

注册账号即表示你同意我们的 使用条款 社区准则。请阅读我们的 隐私政策 ,了解我们如何处理你的信息。