Real-Time Detection of Electronic Components in Waste Printed Circuit Boards: A Transformer-Based Approach
Muhammad Mohsin, Stefano Rovetta, Francesco Masulli, Alberto Cabri
TL;DR
The paper investigates real-time detection and localization of electronic components on waste PCBs using the Real-Time DETR transformer-based detector, comparing its performance against CNN-based CNN detectors such as YOLOv8 and YOLOv9 on a custom V-PCB dataset. The RT-DETR model, including large and edge-oriented variants, demonstrates very high $mAP_{50}$ and low latency, highlighting the viability of transformer-based architectures for edge-enabled, circular-economy applications in CRM extraction. While RT-DETR generally outperforms CNN-based models in key metrics and speed, YOLOv9 can offer marginal gains in $mAP_{50-95}$, underscoring a trade-off between accuracy and resource constraints. Overall, the work supports transformer-based approaches as a strong option for real-time WPCB component recognition to improve selective disassembly and CRM recovery at the edge.
Abstract
Critical Raw Materials (CRMs) such as copper, manganese, gallium, and various rare earths have great importance for the electronic industry. To increase the concentration of individual CRMs and thus make their extraction from Waste Printed Circuit Boards (WPCBs) convenient, we have proposed a practical approach that involves selective disassembling of the different types of electronic components from WPCBs using mechatronic systems guided by artificial vision techniques. In this paper we evaluate the real-time accuracy of electronic component detection and localization of the Real-Time DEtection TRansformer model architecture. Transformers have recently become very popular for the extraordinary results obtained in natural language processing and machine translation. Also in this case, the transformer model achieves very good performances, often superior to those of the latest state of the art object detection and localization models YOLOv8 and YOLOv9.