[1]张晋钊,刘新妹,等.改进 YOLOv5s 的 PCB 元器件检测技术[J].机械与电子,2024,42(10):35-41.
　ZHANG Jinzhao,LIU Xinmei,et al.Improved YOLOv5s Component Detection Technology for PCB[J].Machinery & Electronics,2024,42(10):35-41.
点击复制

改进 YOLOv5s 的 PCB 元器件检测技术()
分享到：

《机械与电子》[ISSN:1001-2257/CN:52-1052/TH]

卷:
42
期数:
2024年10期
页码:
35-41
栏目:
自动控制与检测
出版日期:
2024-10-30

文章信息/Info

Title:
Improved YOLOv5s Component Detection Technology for PCB
文章编号:
1001-2257 ( 2024 ) 10-0035-07
作者:
张晋钊 1 ; 2 ; 刘新妹 1 ;  2 ; 殷俊龄 1 ; 2
1. 中北大学电子测试技术国家重点实验室,山西 太原 030051 ;
2. 中北大学信息与通信工程学院,山西 太原 030051
Author(s):
ZHANG Jinzhao1 ;  2 ;  LIU Xinmei1 ;  2 ;  YIN Junling1 ; 2
( 1.State Key Laboratory of Electronic Testing Technology , North University of China , Taiyuan 030051 , China ;
2.School of Information and Communication Engineering , North University of China , Taiyuan 030051 , China )
关键词:
YOLOv5s ; 印制电路板; 目标检测; 轻量化
Keywords:
YOLOv5s ;  printed circuit board ;  target detection ;  lightweight
分类号:
TP391
文献标志码:
A
摘要:
针对 PCB 元器件检测中硬件运算需求高、检测速度慢等问题,提出一种基于 YOLOv5s 改进的轻量化 PCB 元器件目标检测算法 YOLO-SCGS 。该算法以 ShuffleNetv2 替换主干网络,在 Neck 层引入 CA 注意力机制,使用轻量级卷积方法代替标准卷积,加入 SPD-Conv 重构 CNN 体系结构,代替卷积步长和卷积层,提高检测精度。实验结果表明,所提算法相较原 YOLOv5s 检测速度提升 16.61% ,浮点计算需求减少 85.44% ,模型权重缩小 84.83% ,有更低的运算需求和更优的检测速度,满足轻量化本地部署的要求。
Abstract:
This paper proposes a lightweight PCB component target detection YOLO-SCGS based on YOLOV5s to address the high hardware computational requirements and slow detection speed in PCB component detection technology.This algorithm replaces the backbone network with ShuffleNetv2 , introduces CA attention mechanism in the Neck layer , and designs a cross level partial network GSCSP module for lightweight standard convolution.Finally , SPD Conv is used to reconstruct the CNN architecture , replacing the convolution step size and convolution layer to improve detection accuracy.The experimental results show that compared to the original YOLOv5s , the algorithm proposed in this paper improves detection speed by 16.61% , reduces floating-point computing requirements by 85.44% , and reduces model weight by 84.83%.It has lower computational requirements and better detection speed , meeting the requirements of lightweight local deployment.

参考文献/References:

[ 1 ] 吴亚帅,刘新妹,殷俊龄,等 .基于 DSO 算法的印刷电路板焊点测试路径优化[ J ] .科学技术与工程, 2021 ,21 ( 14 ): 5840-5846.

[ 2 ] 吴一全,赵朗月,苑玉彬,等 . 基于机器视觉的 PCB 缺陷检测算法研究现状及展望[ J ] . 仪器仪表学报, 2022 , 43 ( 8 ): 1-17.
[ 3 ] 李仁鹰,钱慧芳,郭佳豪,等 . 基于 M-YOLOv4 模型的轻量化目标检测算法[ J ] . 国外电子测量技术, 2022 ,41 ( 4 ): 15-21.
[ 4 ] REN S Q , HE K M , GIRSHICK R , et al.Faster R-CNN : towards real-time object detection with region proposal Networks [ J ] .IEEE Transactions on pattern analysis and machine intelligence , 2017 , 39 ( 6 ): 1137-1149.
[ 5 ] SUNKARA R , LUO T.No more strided convolutions or pooling : a new CNN building block for low-resolution inage s and snall pvjects [ C ] ∥Joint European Conference on Machine Learning and Knowledge Discovery in Databases , 2022 : 443-459.
[ 6 ] 窦子豪,刘新妹,殷俊龄,等 . 基于 YOLOv3 的贴装元器件检测技术[ J ] . 电子测量技术, 2021 , 44 ( 13 ): 127-131.
[ 7 ] 郭策 . 基于深度学习的小型电子元器件检测算法研究[ D ] . 天津:河北工业大学,2022.
[ 8 ] LIU X , HU J S , WANG H X , et al.Gaussian-IoU loss : better learning for bounding box regression on PCB component detection [ J ] .Expert systems with applications , 2022 , 190 : 116178.
[ 9 ] LI Q Z , ZHENG Q W , JIANG S X , et al.An improved YOLOv5 based model for automatic PCB defect detection [ J ] .Journal of physics : conference series , 2024 ,2708 ( 1 ): 012017.
[ 10 ] 剌小兆,刘新妹,殷俊龄 . 基于特征匹配的单目相机姿态解算[ J ] . 计算机系统应用, 2023 , 32 ( 1 ): 310-315.
[ 11 ] REDMON J , DIVVILA S , GIRSHICK R , et al.You only look once : unified , real time object detection [ C ] ∥ 2016 IEEE Conference on Computer Vision and Pattern Recognition( CVPR ),2016 : 779-788.
[ 12 ] REDMON J , FARHADI A.YOLO9000 : better , faster , stronger [ C ] ∥2017 IEEE Conference on Computer Vision and Pattern Recognition( CVPR ), 2017 :6517-6525.
[ 13 ] REDMON J , FARHADI A.YOLOv3 : an incremental improvement [ J ] .Computer vision and pattern recognition , 2018 : 02767.
[ 14 ] HOU Q B , ZHOU D Q , FENG J S.Coordinate attention for efficient mobile network design [ C ]// 2021 IEEE / CVF Conference on Computer Vision and Pattern Recognition ( CVPR ), 2021 : 13708-13717.
[ 15 ] YU M , WAN Q , TIAN S L , et al.Equipment identification and localization method based on improved YOLOv5s model for production line [ J ] .Sensors , 2022 , 22 ( 24 ): 10011.
[ 16 ] 朱辉,李海涛,刘岳鑫,等 . 基于改进 YOLOv5s 网络的杆塔相关目标检测方法[ J ] . 电力大数据, 2023 , 26( 5 ): 62-72.
[ 17 ] LIU X D , CHEN B M.An approach to synthesize diverse underwater image dataset [ J ] .Instrumentation , 2019 ( 3 ): 67-75.
[ 18 ] HAN K , WANG Y H , TIAN Q.GhosrtNet : More features from cheap operations [ C ] ∥2020 IEEE / CVF Conference on Computer Vision and Patten Recognition ( CVPR ), 2020 : 1577-1586.
[ 19 ] HOWARD A , SANDLER M , CHEN B , et al.Searching for MobileNetV3 [ C ] ∥2019 IEEE / CVF International Conference on Computer Vision ( ICCV ), 2019 : 1314-1324.
[ 20 ] TAN M X , LE Q V.EfficientNet : rethinking model scaling for convolutional neural networks [ C ] ∥Proceedings of the? 36th International Conference?on Machine Learning , 2019 : 6105-6114.
[ 21 ] 左安全,秦伦明,王悉,等 . 基于改进 DeepLabv3+ 模型的电力线语义分割方法[ J ] . 无线电工程, 2023 , 53( 1 ): 96-104.
[ 22 ] 张明路,郭策,吕晓玲,等 . 改进的轻量化 YOLOv4 用于电子元器件检测[ J ] . 电子测量与仪器学报, 2021 ,35 ( 10 ): 17-23.

备注/Memo

备注/Memo:
收稿日期: 2024-03-15
基金项目:山西省重点研发项目( 201903D121058 )
作者简介:张晋钊 ( 1999- ),男,山西吕梁人,硕士研究生,研究方向为目标检测;刘新妹 ( 1965- ),女,山西太原人,副教授,硕士研究生导师,研究方向为自动化控制与检测、测试计量技术及仪器;殷俊龄 ( 1953- ),男,山西太原人,正高级工程师,研究方向为嵌入式系统。
更新日期/Last Update: 2024-10-30