Trained only on normal —
it still catches defects it has never seen
LINE SURFACE learns the distribution of normal product images and exposes anything different as a pixel-level heatmap. Start with 30 defect samples — or none — and stay protected against novel defect types.
Imaging & capture
Process-tailored lighting and optics maximize defect contrast
Normal-distribution learning
Patch-level embedding distribution modeled from normal images
Anomaly heatmap & score
Where and how much it differs, at pixel level
Verdict · grading · report
Pass/fail + defect grade + MES history
It does not memorize defects. It understands normal.
Collecting thousands of images per defect type collapses the moment a new defect appears. LINE SURFACE goes the other way — learn the distribution of normal, and surface everything outside it as a defect candidate.
Optimize imaging
Lighting angle, wavelength and optics designed per defect family to secure contrast first
Learn normal embeddings
Patch-level normal distribution modeled from a few hundred normal images — no defect labels
Heatmap & score
Distance from the normal distribution rendered as a pixel heatmap — the verdict evidence is visible
Verdict & grading
Heatmap-based pass/fail with automatic defect size, location and severity records
* Real defects collected in operation are fed back into validation sets and the type classifier.
Normal-only training
200–1,000 normal images are enough to start. Never delay a launch waiting for defect data — ideal for new lines and new products.
Evidence you can see
The heatmap shows why it is NG. One visual artifact serves inspector verification, process root-cause work, and customer claim evidence.
Novel-defect coverage
Unseen defect types are still flagged as "different from normal" — breaking the cycle of rebuilding models after mass escapes.
Overkill management
The line between acceptable blemish and true defect differs per plant. Region-wise sensitivity maps and threshold tuning keep overkill controlled.
Use Case Analysis
Coated webs, molded exteriors, continuous sheets — anything with a surface qualifies.
Started with 30 defect samples, on the line in 4 weeks
A new electrode line had only 30 defect samples; "we need thousands" answers had stalled adoption entirely.
The anomaly model was trained on normal coating images only; the 30 defects were used purely for threshold validation.
99.2% accuracy at 1.8% overkill, deployed in four weeks. Never-collected novel defects were caught before first escape.
One appearance standard instead of many opinions
Pass/fail on scratches, dents and paint blemishes varied by inspector, driving both customer claims and overkill waste.
Sensitivity maps calibrated to limit samples, with heatmap evidence attached to every verdict, formed a standard inspection scheme.
Criteria unified as data: claim responses got faster, and good parts previously scrapped as overkill were recovered.
Full-width, full-length monitoring with line-scan
Pinholes, particles and streaks on high-speed sheet were managed by eye and sampling, producing roll-level losses.
Line-scan full-width imaging combined with anomaly heatmaps, logging each defect position in meters along the web.
Defect positions form a roll map, so downstream removes only affected segments — whole-roll scrap became partial loss.
* Figures and scenarios reflect representative deployments and may vary by process and data conditions.
Specs & Integration
Frequently Asked Questions
Does it really work without defect samples?
We worry about overkill (good parts rejected).
What happens when the product changes?
How does it relate to our existing AOI?
Send 100 sample images — we will show you what is possible
Tell us about your process and data environment — we will come back with a feasibility summary and expected impact.