Deep Learning-Powered Computer Vision System for Selective Disassembly of Waste Printed Circuit Boards

Abstract

The exponential growth of electronic waste is a direct result of nowadays fast technological progress. European Union directives prioritize resource optimization, particularly the circular utilization of Critical Raw Materials (CRMs) present in electronic devices. In our study, we introduce an advanced computer vision system based on the deep learning model YOLOv9, designed to support the robotic selective disassembly of Waste Printed Circuit Boards (WPCBs). This is an effective approach for enhancing the density of specific CRMs and making their extraction more efficient. Our approach leverages chemical-physical processes to efficiently extract CRMs from electronic components. By utilizing distinctive features, we classify these components based on their recyclability, thereby enhancing recycling efforts.