Virtual Mines -- Component-level recycling of printed circuit boards using deep learning
Muhammad Mohsin, Stefano Rovetta, Francesco Masulli, Alberto Cabri
TL;DR
The paper addresses the escalating problem of electronic waste by targeting component-level recycling of printed circuit boards (PCBs) within the circular economy, anchored in the virtual mines concept. It presents a deep learning–driven vision pipeline that employs YOLOv5 to detect PCB components, trained and evaluated on a locally collected V-PCB dataset with ROI-based preprocessing and edge-device inference. Key contributions include the V-PCB dataset, an annotation workflow with eight PCB component classes, and an end-to-end methodology validated on edge hardware, achieving performance metrics such as precision $P = \frac{TP}{TP+FP}$ and recall $R = \frac{TP}{TP+FN}$ with observed values around $P \approx 0.80$ and $R \approx 0.56$ along with standard mAP evaluation. The work demonstrates the feasibility of using AI-driven sorting to enhance material recovery in e-waste recycling, with future directions toward hierarchical classification and OCR for extracting material data from PCBs.
Abstract
This contribution gives an overview of an ongoing project using machine learning and computer vision components for improving the electronic waste recycling process. In circular economy, the "virtual mines" concept refers to production cycles where interesting raw materials are reclaimed in an efficient and cost-effective manner from end-of-life items. In particular, the growth of e-waste, due to the increasingly shorter life cycle of hi-tech goods, is a global problem. In this paper, we describe a pipeline based on deep learning model to recycle printed circuit boards at the component level. A pre-trained YOLOv5 model is used to analyze the results of the locally developed dataset. With a different distribution of class instances, YOLOv5 managed to achieve satisfactory precision and recall, with the ability to optimize with large component instances.