GermanyManufacturing2024 – ongoingWieland Group
Deep-learning inspection of copper component surfaces
Wieland tested AI inspection of reflective copper components under practical production conditions
Problem
Manual inspection took substantial effort. Round components and reflective copper surfaces made visual defect detection difficult.
Approach
Wieland and Capgemini developed a deep-learning inspection system. They expanded scarce training examples through data augmentation and tested recognition on edge hardware under practical conditions. Cloud management prepared the system for wider deployment.
Results
No published figures
Takeaway
Capgemini reported reliable defect recognition under practical conditions after expanding scarce training data. Wider deployment remained for Wieland to carry out.
Categories
- Industry
- Manufacturing
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Computer vision
- Adoption
- Partner co-development
- Evidence
- Vendor case study
Sources
Source summaries
Capgemini reported successful feasibility and practical-condition testing of copper-surface inspection. Its cloud-managed system prepared Wieland to extend inspection to additional cells and products; completed deployment across those locations was not established.
Full details are in the original sources.