The problem
Manual inspection is inconsistent, and automated systems are too expensive for small factories.
What to build
Build a camera station that learns what good looks like from a few examples and flags defects.
Success looks like
Defect catch rate above 90% on a sample batch with a tuneable false-reject rate.
Suggested technologies
- Anomaly detection
- Few-shot learning
- Industrial cameras
- Edge compute
- Operator UI
An original, independent statement. It is not an official Smart India Hackathon problem statement, and Arventiq Labs is not affiliated with or endorsed by that programme. The scenario is illustrative and does not describe any specific organisation. Free to use with credit.