What Percentage Is AI Inspection? The Rates – Defect Clinic EP.10

Good housings rejected in auto parts inspection? "What percentage" is answered by the false-reject and missed-defect rates and their validation set.

English audio: choose it in the player under ⚙ › Audio track. Captions are under CC.

EP.10 · False rejects and misses · Machine Vision False Rejects on Auto Parts: How to Read the Numbers

Good machined aluminium housings come out NG (false rejects). "So what percentage is the AI?"

Glare and tool marks on machined faces get rejected, and re-checks cost time. The answer is not one number: it is the two rates and the validation set.

Both rates together, on real samples.

Chapters: 0:00 So, what percentage? / 0:22 Why false rejects happen: glare and lighting / 1:02 Reading the numbers: false rejects vs escapes / 1:21 Actual screens: the IntLearn Pro demo / 2:04 Test on real samples, then run it in IntSight / 2:25 Your 5 line conditions

Can I apply the fix from a video to my line as is? Each video is a short animation that walks through a problem and a fix. Whether it works on your line depends on a review of samples, imaging conditions and the process.

New episodes go up on YouTube when ready, each summarized here.

* Season 1 is an animated dramatization. CEO Na's voice is an AI clone of the CEO's own, used with consent; other voices are AI-synthesized and the music is AI-generated. On-screen values, charts and digital-twin scenes are examples; defect photos and heatmaps are inspection examples. IntLearn ML autonomous control is under development and needs on-site validation.