基于改进YOLO的印刷电路板表面缺陷检测算法

打开文本图片集
关键词:图像处理;PCB缺陷检测;YOLO;多尺度卷积注意力机制;高效特征金字塔网络中图分类号:TP39 文献标识码:A doi:0.3788/CJLCD.05-037 CSTR:37.4.CJLCD.05-0
Surface defect detection on printed circuit boards based on improved YOL01
YANG Yanping1.², GAO Junwei 1,2* , LIU Zhaolong 1,2 , XING Rongxin 1,2
(1.School of Automation,Qingdao University,Qingdao 266O71,China; 2. Shandong Key Laboratory ofIndustrial Control Technology,Qingdao 266O71, China)
Abstract: To address the challenges of low accuracy,false detections,and missed detections in PCB defect inspection under scenarios with smalldefect sizes,complex circuit backgrounds,and irregular defect shapes, this paper proposes a surface defect detection method based on an improved YOLOvl1 architecture.First,a multi-scale convolutional attention (MSCA)mechanism is integrated into the backbone to enhance the model's sensitivity to subtle defects.Second,the neck structure is replaced with the high-level screening feature pyramid network (HS-FPN),which improves defect detection across sizes by using channel atention(CA) and selective feature fusion (SFF).Third,the original C3K2 module is enhanced by combining ConvFormer with a convolutional gated linear unit (CGLU),which improves feature representation while reducing computational cost. Furthermore,the unified-IoU(UIoU)lossfunction is applied to dynamically adjust the weighting ofcandidate boxes,enhancing both localization accuracy and convergence speed. Extensive experiments are conducted on a self-constructed PCB dataset with six defect categories. The proposed model achieves a mean average precision (mAP) of 89.1% , surpassing the baseline YOLOvl1 by 5.3% in mAP. In addition, the precision and recall are improved by (20 2.3% and 6.1% ,respectively,while the model's parameter count is reduced by 28.3% . The results demonstrate the efectiveness and practical potential of the proposed approach for real-world PCB defect detection tasks.
Key words: image processing;PCB defect detection;YOLOll;multi-scale convolutional atention; HS-FPN
1引言
印刷电路板(PrintedCircuitBoard,PCB)是电子设备的基础组成部分,广泛应用于通讯、计算机、汽车电子、医疗设备及工业自动化等多个领域。(剩余16014字)