ITP Aero – Automatic X-Ray Defect Detection
ITP Aero
Sector: Industry
Business Case
X-ray-based inspection processes are very relevant for detecting quality problems. The aim is to build a system capable of assisting inspectors in the detection of defects present in the parts.
Objectives
Improve capability (sizes detected, probability) and efficiency (time required) in defect detection.
Use case
An application is built to generate labelled datasets. Once the dataset is generated, AI models will be generated to help in the detection of defects starting with the most common types.
Infrastructure
On Premise
Technology
Computer vision Machine learning and deep learning
Data
Images Private dataset
Resources
IT staff (2), technology team (1), inspector team (3) & data science team (1).
Difficulties and learning
Resistance to change, number of images required, integration with corporate systems.
KPIs (business impact and metrics of the model)
Improved POD curves.
Funding
–
Collaborators, Partners
UC3M
