Printed circuit boards (PCBs) from waste electrical equipment are a valuable source of materials, containing raw materials such as gold, copper and palladium.
Consumer electronics and information and telecommunications technology are presenting the recycling industry with a problem: there is a lack of suitable methods for determining the exact value of a large volume of printed circuit boards with high precision and high throughput.
The goal of the PCBcycle project is to carry out a complete online evaluation of the PCBs followed immediately afterward by sorting. This results in a system and process for the automatic sorting of waste printed circuit boards (WPCBs). On the basis of the predicted value, it can be determined whether it is financially worthwhile to extract valuable components.
The system records dual-energy X-ray images of the PCBs on a conveyor belt. These images are pre-processed and fed into a deep neural network that recognizes individual components. The system calculates the value of each component in a PCB. An important advantage of X-ray images here is that they allow components to be recognized at the front and back simultaneously.