Major project
Manufacturing, Robotics & Industry 4.0
Defect detection on the edge
Build a defect model and deploy it to a Raspberry Pi or similar device next to the line, with monitoring.
The brief
Factories often cannot send images to the cloud, and the line cannot wait. Build a defect model, shrink it for an edge device, deploy it next to the line, and monitor its accuracy and speed over time. Good looks like: a model that meets a speed target on the device, an accuracy comparison before and after shrinking, and a monitoring dashboard.
What you will build
- Train a defect detection model
- Export to ONNX and quantise it
- Deploy on a Raspberry Pi or similar device
- Measure speed and accuracy on the device
- Send results and health checks to a dashboard
- Plan how to update the model on the device
Free, open data to build it with
Check each dataset's licence before you publish your work. Your mentor gives your team its own version of the brief (data, scope or region), so every team's project is its own.