PCB-Vision: A Multiscene RGB-Hyperspectral Benchmark Dataset of Printed Circuit Boards

TL;DR

PCB-Vision addresses the need for non-invasive PCB analysis in e-waste recycling by introducing a large, open RGB–HSI benchmark dataset comprising 53 PCBs with 53 RGB images and 224-band VNIR hyperspectral cubes, along with General and Monoseg ground truths for IC, Capacitor, and Connectors. The paper benchmarks five semantic segmentation models (Unet, ResUnet, Attention Unet, DeepLabv3+, LinkNet) on both data modalities, employing data augmentation, class weighting, and PCB-background masking to mitigate background interference. Key contributions include dataset acquisition and annotation protocols, statistical characterizations, and a comprehensive set of baseline results that reveal modality-specific strengths and limitations, especially regarding spectral-spatial information in HS data. The work advances open benchmarking for PCB analysis, supports SDG-aligned recycling initiatives, and invites community-driven development of robust, generalizable hyperspectral PCB processing pipelines.

Abstract

Addressing the critical theme of recycling electronic waste (E-waste), this contribution is dedicated to developing advanced automated data processing pipelines as a basis for decision-making and process control. Aligning with the broader goals of the circular economy and the United Nations (UN) Sustainable Development Goals (SDG), our work leverages non-invasive analysis methods utilizing RGB and hyperspectral imaging data to provide both quantitative and qualitative insights into the E-waste stream composition for optimizing recycling efficiency. In this paper, we introduce 'PCB-Vision'; a pioneering RGB-hyperspectral printed circuit board (PCB) benchmark dataset, comprising 53 RGB images of high spatial resolution paired with their corresponding high spectral resolution hyperspectral data cubes in the visible and near-infrared (VNIR) range. Grounded in open science principles, our dataset provides a comprehensive resource for researchers through high-quality ground truths, focusing on three primary PCB components: integrated circuits (IC), capacitors, and connectors. We provide extensive statistical investigations on the proposed dataset together with the performance of several state-of-the-art (SOTA) models, including U-Net, Attention U-Net, Residual U-Net, LinkNet, and DeepLabv3+. By openly sharing this multi-scene benchmark dataset along with the baseline codes, we hope to foster transparent, traceable, and comparable developments of advanced data processing across various scientific communities, including, but not limited to, computer vision and remote sensing. Emphasizing our commitment to supporting a collaborative and inclusive scientific community, all materials, including code, data, ground truth, and masks, will be accessible at https://github.com/hifexplo/PCBVision.

Figures (11)

• Figure 1: PCB-Vision setup to results: (a) RGB images and HS data cubes are acquired, (b) data normalization and preprocessing, (c) data preparation for ML model pipeline, (d) segmentation results
• Figure 2: Acquisition setup at Helios Lab sudharshan2020object
• Figure 3: Four spectra from PCB 1 HSI were captured from randomly selected points on the surface of our classes of interest, along with the spectra from the conveyor belt background. 'Conveyor belt' (orange), 'IC' (red), 'Capacitor' (green), and 'Connectors' (blue).
• Figure 4: 'General' and 'Monoseg' segmentation ground truth projected on PCB1 RGB image. 'IC' (red), 'Capacitor' (green), 'Connectors' (blue).
• Figure 5: 'General' VS 'Monoseg' segmentation ground truth of PCB 1 RGB image. 'IC' (red), 'Capacitor' (green), 'Connectors' (blue).