An end-to-end asbestos detection pipeline, with CVAT as its annotation backbone.
Read the story- asbestos fibers annotated
- ~100,000
- production model in customer labs
- Deployed
CVAT for Manufacturing & Industrial AI
CVAT turns line-camera, inspection, and microscope imagery into validated training data for defect detection, quality control, and assembly verification, trusted by teams like ProMetronics.
Use cases
Turn Production Imagery Into Fewer Escaped Defects
Surface defect detection
Mark scratches, dents, and cracks on parts as they leave the line.
PCB and electronics inspection
Find missing, shifted, and bridged components on every board.
Assembly verification
Confirm every bolt, connector, and bracket is in place before it ships.
Microscopy and particle analysis
Count fibers and particles in SEM and optical microscope images.
Worker safety and PPE
Detect hard hats, vests, and restricted-zone entries on the floor.
How it Works
Integrate Seamlessly
Start labeling without changing how you work
Connect your line-camera archive, existing labels, and review stages as they are. Nothing to re-plumb.
Book a demoProjects/line-3-inspection
Search sourcesAdd source
SourceStatus
1s3://line-cameras/station-07Syncing…
2gs://aoi-images/pcb-2026Syncing…
3azure://sem-scans/lab-01Syncing…
4labels_v3 · COCO 1.0Syncing…
5Annotation → Validation → AcceptanceSyncing…
6webhook update:job → trainingSyncing…
7defect_catalog_v5.pdfSyncing…
Accelerate Validation
Clear labeling backlogs faster
AI pre-labels every part image. Your team or our experts fix what it misses.
Book a demoJobs/Part 20417
housing #1
scratch #2
burr #3
bolt_hole #4
pore #5
Ship with Confidence
Catch label errors before they reach training
Every job is scored against ground truth, so only clean data reaches training.
Book a demoTasks/Quality control
Mean annotation quality
98.4%
Target 95%
Honeypots
CVAT is how ML teams ship industrial AI.
