Research on PCB defect detection algorithm based on LPCB-YOLO

IntroductionIn response to the challenges of small target size, slow detection speed, and large model parameters in PCB surface defect detection, LPCB-YOLO was designed. The goal was to ensure detection accuracy and comprehensiveness while significantly reducing model parameters and improving comput...

Full description

Saved in:
Bibliographic Details
Main Authors: Haiyan Zhang, Yazhou Li, Dipu Md Sharid Kayes, Zhaoyu Song, Yuanyuan Wang
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-01-01
Series:Frontiers in Physics
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fphy.2024.1472584/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1841561205993373696
author Haiyan Zhang
Haiyan Zhang
Yazhou Li
Yazhou Li
Dipu Md Sharid Kayes
Dipu Md Sharid Kayes
Zhaoyu Song
Zhaoyu Song
Yuanyuan Wang
Yuanyuan Wang
author_facet Haiyan Zhang
Haiyan Zhang
Yazhou Li
Yazhou Li
Dipu Md Sharid Kayes
Dipu Md Sharid Kayes
Zhaoyu Song
Zhaoyu Song
Yuanyuan Wang
Yuanyuan Wang
author_sort Haiyan Zhang
collection DOAJ
description IntroductionIn response to the challenges of small target size, slow detection speed, and large model parameters in PCB surface defect detection, LPCB-YOLO was designed. The goal was to ensure detection accuracy and comprehensiveness while significantly reducing model parameters and improving computational speed.MethodFirst, the feature extraction networks consist of multiple CSPELAN modules for feature extraction of small target defects on PCBs. This allows for sufficient feature representation while greatly reducing the number of model parameters. Second, the C-SPPF module enables the fusion of high-level semantic expression with low-level feature layers to enhance global feature perception capability, improving the overall contextual expression of the backbone and thereby enhancing model performance. Finally, the C2f-GS module is designed to fuse high-level semantic features and low-level detail features to enhance the feature representation capability and model performance.ResultsThe experimental results show that the LPCB-YOLO model reduces the model size by 24% compared to that of the YOLOv8 model while maintaining high precision and recall at 97.0%.
format Article
id doaj-art-ab65d8977bcb4656ba2d95ac0f8014fb
institution Kabale University
issn 2296-424X
language English
publishDate 2025-01-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Physics
spelling doaj-art-ab65d8977bcb4656ba2d95ac0f8014fb2025-01-03T05:10:22ZengFrontiers Media S.A.Frontiers in Physics2296-424X2025-01-011210.3389/fphy.2024.14725841472584Research on PCB defect detection algorithm based on LPCB-YOLOHaiyan Zhang0Haiyan Zhang1Yazhou Li2Yazhou Li3Dipu Md Sharid Kayes4Dipu Md Sharid Kayes5Zhaoyu Song6Zhaoyu Song7Yuanyuan Wang8Yuanyuan Wang9College of Computer and Software Engineering, Huaiyin Institute of Technology, Huaian, ChinaLaboratory for Internet of Things and Mobile Internet Technology of Jiangsu Province, Huaian, ChinaCollege of Computer and Software Engineering, Huaiyin Institute of Technology, Huaian, ChinaLaboratory for Internet of Things and Mobile Internet Technology of Jiangsu Province, Huaian, ChinaCollege of Computer and Software Engineering, Huaiyin Institute of Technology, Huaian, ChinaLaboratory for Internet of Things and Mobile Internet Technology of Jiangsu Province, Huaian, ChinaCollege of Computer and Software Engineering, Huaiyin Institute of Technology, Huaian, ChinaLaboratory for Internet of Things and Mobile Internet Technology of Jiangsu Province, Huaian, ChinaCollege of Computer and Software Engineering, Huaiyin Institute of Technology, Huaian, ChinaLaboratory for Internet of Things and Mobile Internet Technology of Jiangsu Province, Huaian, ChinaIntroductionIn response to the challenges of small target size, slow detection speed, and large model parameters in PCB surface defect detection, LPCB-YOLO was designed. The goal was to ensure detection accuracy and comprehensiveness while significantly reducing model parameters and improving computational speed.MethodFirst, the feature extraction networks consist of multiple CSPELAN modules for feature extraction of small target defects on PCBs. This allows for sufficient feature representation while greatly reducing the number of model parameters. Second, the C-SPPF module enables the fusion of high-level semantic expression with low-level feature layers to enhance global feature perception capability, improving the overall contextual expression of the backbone and thereby enhancing model performance. Finally, the C2f-GS module is designed to fuse high-level semantic features and low-level detail features to enhance the feature representation capability and model performance.ResultsThe experimental results show that the LPCB-YOLO model reduces the model size by 24% compared to that of the YOLOv8 model while maintaining high precision and recall at 97.0%.https://www.frontiersin.org/articles/10.3389/fphy.2024.1472584/fullprinted circuit boardlightweight networkELANdefects detectiontiny target detection
spellingShingle Haiyan Zhang
Haiyan Zhang
Yazhou Li
Yazhou Li
Dipu Md Sharid Kayes
Dipu Md Sharid Kayes
Zhaoyu Song
Zhaoyu Song
Yuanyuan Wang
Yuanyuan Wang
Research on PCB defect detection algorithm based on LPCB-YOLO
Frontiers in Physics
printed circuit board
lightweight network
ELAN
defects detection
tiny target detection
title Research on PCB defect detection algorithm based on LPCB-YOLO
title_full Research on PCB defect detection algorithm based on LPCB-YOLO
title_fullStr Research on PCB defect detection algorithm based on LPCB-YOLO
title_full_unstemmed Research on PCB defect detection algorithm based on LPCB-YOLO
title_short Research on PCB defect detection algorithm based on LPCB-YOLO
title_sort research on pcb defect detection algorithm based on lpcb yolo
topic printed circuit board
lightweight network
ELAN
defects detection
tiny target detection
url https://www.frontiersin.org/articles/10.3389/fphy.2024.1472584/full
work_keys_str_mv AT haiyanzhang researchonpcbdefectdetectionalgorithmbasedonlpcbyolo
AT haiyanzhang researchonpcbdefectdetectionalgorithmbasedonlpcbyolo
AT yazhouli researchonpcbdefectdetectionalgorithmbasedonlpcbyolo
AT yazhouli researchonpcbdefectdetectionalgorithmbasedonlpcbyolo
AT dipumdsharidkayes researchonpcbdefectdetectionalgorithmbasedonlpcbyolo
AT dipumdsharidkayes researchonpcbdefectdetectionalgorithmbasedonlpcbyolo
AT zhaoyusong researchonpcbdefectdetectionalgorithmbasedonlpcbyolo
AT zhaoyusong researchonpcbdefectdetectionalgorithmbasedonlpcbyolo
AT yuanyuanwang researchonpcbdefectdetectionalgorithmbasedonlpcbyolo
AT yuanyuanwang researchonpcbdefectdetectionalgorithmbasedonlpcbyolo