EP-YOLO：一种集成了坐标注意力和多级梯度流优化的印刷电路板缺陷检测网络
EP-YOLO: A Printed Circuit Board Defect Detection Network Integrating Coordinate Attention and Multi-Level Gradient Flow Optimization.
摘要(中文译文)
电子产品对印刷电路板（PCB）的依赖日益增强。近年来，PCB缺陷检测技术已取得相当成果，但缺陷目标小、背景复杂、集成元件多，给产品质量控制带来了巨大挑战。因此，本文提出了一种融合坐标注意力与多级梯度流优化的印刷电路板缺陷检测网络，称为EP-YOLO，以提高印刷电路板缺陷的检测精度。首先，本文通过重构小目标检测头，提升模型捕捉小目标细节的能力。在此基础上，设计了浅层上下文特征提取（SCFE）网络，将浅层特征与多尺度特征融合，有效防止浅层细节信息和纹理特征的丢失。同时，为抑制背景噪声，本文设计了融合坐标注意力的多级特征梯度流优化模块（C2f_CA，简称CCA）以及跨阶段部分频域全核模块（CSPFOK），以增强特征提取能力。最后，引入了SCYLLA-IoU（SIoU）优化模型训练过程。实验结果表明，EP-YOLO在PKU-Market-PCB数据集上达到97.5%的mAP50，在DeepPCB数据集上达到98.5%的mAP50，参数量减少约12.55%，优于主流目标检测网络。结果凸显了EP-YOLO的潜在能力，为工业PCB缺陷检测提供了一种强大而有效的解决方案。
Abstract
The dependence of electronic products on printed circuit boards (PCBs) is increasing. In recent years, PCB defect detection technology has achieved considerable results, but the defect targets are small, the background is complex, and there are many integrated components, which poses great challenges to product quality control. Therefore, this paper proposes a printed circuit board defect detection network that integrates coordinate attention and multi-level gradient flow optimization, called EP-YOLO, to improve the detection accuracy of printed circuit board defects. Firstly, this paper improves the model's ability to capture small target details by reconstructing the detection head of the small target. Based on this, a Shallow Context Feature Extraction (SCFE) network is designed to fuse shallow features with multi-scale features, effectively preventing the loss of shallow detail information and texture features. At the same time, in order to suppress background noise, this paper designs a multi-level feature gradient flow optimization module (C2f_CA, abbreviated as CCA) that integrates coordinate attention and a Cross-Stage Partial Frequency-domain Omni-Kernel module (CSPFOK) to enhance feature extraction capability. Finally, the SCYLLA-IoU (SIoU) optimization model training process was introduced. The experimental results showed that EP-YOLO achieved 97.5% mAP50 on the PKU-Market-PCB dataset and 98.5% mAP50 on the DeepPCB dataset, with a parameter reduction of approximately 12.55%, outperforming popular object detection networks. The results highlight the potential capabilities of EP-YOLO, providing a powerful and effective solution for industrial PCB defect detection.
如何引用
AMA
Xiangsuo Fan, Can Yang, Ling Yu. EP-YOLO: A Printed Circuit Board Defect Detection Network Integrating Coordinate Attention and Multi-Level Gradient Flow Optimization.. Sensors (Basel, Switzerland). 2026; doi:10.3390/s26103106.
APA
Xiangsuo Fan, Can Yang, Ling Yu (2026). EP-YOLO: A Printed Circuit Board Defect Detection Network Integrating Coordinate Attention and Multi-Level Gradient Flow Optimization.. Sensors (Basel, Switzerland). https://doi.org/10.3390/s26103106
GB/T 7714
Xiangsuo Fan, Can Yang, Ling Yu. EP-YOLO: A Printed Circuit Board Defect Detection Network Integrating Coordinate Attention and Multi-Level Gradient Flow Optimization.[J]. Sensors (Basel, Switzerland), 2026 doi:10.3390/s26103106.
在 PubMed增强版 阅读全文
翻译全文 · AI 文献问答 · PDF 原文下载 · 引用导出