2026-10-14T18:06:25Z https://www.scidb.cn/oai

10.57760/sciencedb.44528 2026-07-30T11:09:56Z

2026-07-30 A synthetic CT dataset for beam‑hardening artifact correction in printed circuit board assembly imaging doi:10.57760/sciencedb.44528 en Printed circuit board assembly (PCBA) reverse engineering is essential for schematic recovery, design verification, defect assessment, and maintenance of electronic systems. However, metal-induced beam-hardening artifacts in PCBA computed tomography (CT) imaging severely obscure internal components and interconnections, while the scarcity of paired artifact and ground-truth data limits deep learning-based beam-hardening correction. To address this gap, we propose a design-driven framework for generating physically realistic PCBA CT datasets directly from open-source PCB design files. The PCB projects are parsed and converted into multilayer volumetric models containing air, copper, solder, and FR4. Metallic components and vias are further modeled to construct synthetic digital PCBA phantoms consistent with practical electrical connectivity and layer-stackup rules. An efficient physics-based simulation pipeline is then developed to generate paired artifact-affected and artifact-free CT images. Comparisons with Monte Carlo simulations validate the realism and computational efficiency of the proposed method. For technical validation, we design both numerical simulation and real-data experiments to comprehensively evaluate the dataset. Specifically, we partition the 200 volumetric datasets into 174,798 non-overlapping 512×512 2D CT slices, and then randomly selected 1,596 paired artifact and ground-truth images for model training and 294 artifact images for testing. Three representative DL-based methods (U‑Net, TransNet, and β‑CycleGAN) are trained and evaluated on this dataset. Both qualitative and quantitative results demonstrate that our dataset consistently enables effective BHC across all three models, and further cross‑dataset experiments on real measured data confirm its reasonable generalization capability. Beam‑hardening artifacts; PCBA CT imaging; Synthetic dataset Xinrui Zhang Junru Ren Ailong Cai Zhizhong Zheng Ningning Liang Lei Li Bin Yan PUBLIC https://creativecommons.org/licenses/by-nc/4.0/ dataset Science Data Bank