In factories, because a large number of products are produced every day, it’s hard to control the quality of every item to ensure that defective products are not shipped. This is often accomplished manually by workers on the assembly line. We can automate this process by developing an AI system that can detect faulty items from images captured by cameras and send notifications for further actions.
- Data: images that are taken by cameras, including normal and faulty samples.
- Model: we can train a model that can classify images into various categories: normal, faulty, uncertain. From the output of the model, we can take appropriate actions.