At Cegeka, we are excited about data analysis. We help our clients to gain valuable insights from data. Insights that enable them to make important decisions. Would you like to dive into data lakes with us?
For a waste processing company we created an AI system that can recognize dirt in bales of compressed PET bottles with the use of a camera. In this way they can intervene when it’s necessary. If the camera sees something that doesn't belong in such a bale, an alert is triggered so that the unwanted parts can be taken out. Moreover, this information is added on their data platform to the information for their clients. In this way they know what kind of quality is the plastic they are getting, how it can be used, and so much more.
We're actually going to train the AI system with pictures of pet bales. And each picture we're going to give a label: dirty or clean. We call that ‘labeling’. In that way the system can 'learn' what is dirt and what is plastic. The more pictures we provide and the more variation in those pictures, the faster the AI system learns and the faster it will be able to make decisions on its own.