Research on YOLO algorithm for lightweight PCB defect detection based on MobileViT

Abstract

Current you only look once (YOLO)-based algorithm model is facing the challenge of overwhelming parameters and calculation complexity under the printed circuit board (PCB) defect detection application scenario. In order to solve this problem, we propose a new method, which combined the lightweight network mobile vision transformer (MobileViT) with the convolutional block attention module (CBAM) mechanism and the new regression loss function. This method needed less computation resources, making it more suitable for embedded edge detection devices. Meanwhile, the new loss function improved the positioning accuracy of the bounding box and enhanced the robustness of the model. In addition, experiments on public datasets demonstrate that the improved model achieves an average accuracy of 87.9% across six typical defect detection tasks, while reducing computational costs by nearly 90%. It significantly reduces the model’s computational requirements while maintaining accuracy, ensuring reliable performance for edge deployment.

Keywords

A

References

[1] WuX C, MiaoC W, WangY B, et al.. Visual defect detection method for shaft tile parts based on the I-YOLO model. Automation applications, 2024, 65(04): 172-175[J]

[2] ZhuCResearch on circuit board appearance defect diagnosis based on deep learning, 2023, Dalian. Dalian Jiaotong University. [D]

[3] MaCDesign and research of PCB defect detection system based on image processing, 2009, Chengdu. Southwest Jiaotong University. [D]

[4] HuW YResearch on automatic PCB defect detection system based on machine vision, 2019, Guangzhou. Guangdong University of Technology. [D]

[5] FangQ, TuZ Y, XiangM Y, et al.. Research on PCB defect detection based on lightweight YOLOv5. Internet of things technology, 2023, 13(8): 15-19[J]

[6] WangS Q, ZhangZ Y, ZhuW X, et al.. PCB surface defect detection based on improved YOLOv5. Instrument technology and sensors, 2023, 5: 106-111[J]

[7] XuS, ChenZ, ZhangH, et al.. Improved remote sensing image target detection based on YOLOv7. Optoelectronics letters, 2024, 20(4): 234-242[J]

[8] YangM, ChenG, LvF, et al.. Detection of early bruises in apples using hyperspectral imaging and an improved MobileViT network. Journal of food science, 2024, 89(12): 8581-8593[J]

[9] ChenH, ZhouG, HeW, et al.. Classification and identification of agricultural products based on improved MobileNetV2. Scientific reports, 2024, 1413454[J]

[10] LiuP, XingW, MaY. MCR-YOLO model for underwater target detection based on multi-color spatial features. Optoelectronics letters, 2024, 20(5): 313-320[J]

[11] WangW, TanX, ZhangP, et al.. A CBAM based multiscale transformer fusion approach for remote sensing image change detection. IEEE journal of selected topics in applied earth observations and remote sensing, 2022, 15: 6817-6825[J]

[12] ZhouY, YuanM, ZhangJ, et al.. Review of vision-based defect detection research and its perspectives for printed circuit board. Journal of manufacturing systems, 2023, 70: 557-578[J]

RIGHTS & PERMISSIONS

Tianjin University of Technology