机电工程技术2025，Vol.54Issue(6)：64-69,6.DOI:10.3969/j.issn.1009-9492.2025.06.012
基于改进YOLOv8的PCB缺陷检测算法
PCB Defect Detection Algorithm Based on Improved YOLOv8
孔祥强 1刘广敏 1高彦臣2
作者信息
- 1. 山东交通学院轨道交通学院,济南 250357
- 2. 青岛智能产业技术研究院,山东 青岛 266114
- 折叠
摘要
Abstract
Printed circuit board(PCB)is an important part of electronic products,and its quality is the key to the normal use of electronic products.In order to improve the detection accuracy of the model for PCB small target defects,a PCB defect detection algorithm YOLOv8-G is proposed based on the improved YOLOv8.Firstly,the SE channel attention mechanism is introduced into the necking network to optimise the feature extraction capability for small target defects and improve the detection accuracy of the model for small target defects.Second,the weighted bidirectional feature pyramid network BiFPN structure is used to replace the original PANet structure in the neck network to enhance the multi-scale feature fusion capability of the model and make it more adaptable to the detection needs of small target defects.Finally,the WIOU loss function is used to replace the traditional CIOU loss function to improve the detection accuracy and robustness of the model.The experimental results show that the model proposed achieves 93.1%,87.4%and 93.1%in precision,recall and average detection accuracy,respectively,which are 2.9%,3.4%and 2.1%higher than the original model.It effectively improves the detection precision of PCB defects and enhances its stability and reliability in practical industrial production applications.关键词
PCB/缺陷检测/YOLOv8/注意力机制/损失函数Key words
PCB/defect detection/YOLOv8/attention mechanism/loss function分类
计算机与自动化引用本文复制引用
孔祥强,刘广敏,高彦臣..基于改进YOLOv8的PCB缺陷检测算法[J].机电工程技术,2025,54(6):64-69,6.
