TY - JOUR
T1 - Multiple Detection Model Fusion Framework for Printed Circuit Board Defect Detection
AU - Wu, Xing
AU - Zhang, Qingfeng
AU - Wang, Jianjia
AU - Yao, Junfeng
AU - Guo, Yike
N1 - Funding Information:
The Natural Science Foundation of Shanghai (No. 20ZR1420400), and the State Key Program of National Natural Science Foundation of China (No. 61936001)
Publisher Copyright:
© Shanghai Jiao Tong University 2022
PY - 2023/12
Y1 - 2023/12
N2 - The printed circuit board (PCB) is an indispensable component of electronic products, which determines the quality of these products. With the development and advancement of manufacturing technology, the layout and structure of PCB are getting complicated. However, there are few effective and accurate PCB defect detection methods. There are high requirements for the accuracy of PCB defect detection in the actual production environment, so we propose two PCB defect detection frameworks with multiple model fusion including the defect detection by multi-model voting method (DDMV) and the defect detection by multi-model learning method (DDML). With the purpose of reducing wrong and missing detection, the DDMV and DDML integrate multiple defect detection networks with different fusion strategies. The effectiveness and accuracy of the proposed framework are verified with extensive experiments on two open-source PCB datasets. The experimental results demonstrate that the proposed DDMV and DDML are better than any other individual state-of-the-art PCB defect detection model in F1-score, and the area under curve value of DDML is also higher than that of any other individual detection model. Furthermore, compared with DDMV, the DDML with an automatic machine learning method achieves the best performance in PCB defect detection, and the F1-score on the two datasets can reach 99.7% and 95.6% respectively.印刷电路板(PCB)是电子产品不可或缺的组成部分,它决定了这些产品的质量。随着制造技术的发展和进步,PCB的布局和结构变得越来越复杂。然而,有效准确的PCB缺陷检测方法却很少。在实际生产环境中,对PCB缺陷检测的准确性有很高的要求,因此我们提出了两种包括多模型融合的PCB缺陷检测框架,包括多模型投票方法(DDMV)和多模型学习方法(DDML)的缺陷检测。为了减少错误和漏检,DDMV和DDML将多个具有不同融合策略的缺陷检测网络进行整合。通过对两个开源PCB数据集进行大量实验,验证了所提出框架的有效性和准确性。实验结果表明,所提出的DDMV和DDML在F1分数方面优于其他任何单独的最先进的PCB缺陷检测模型,而DDML的曲线下面积值也高于其他任何单独的检测模型。此外,与DDMV相比,使用自动机器学习方法的DDML在PCB缺陷检测方面达到了最好的性能,在两个数据集上的F1分数分别可达到99.7%和95.6%。
AB - The printed circuit board (PCB) is an indispensable component of electronic products, which determines the quality of these products. With the development and advancement of manufacturing technology, the layout and structure of PCB are getting complicated. However, there are few effective and accurate PCB defect detection methods. There are high requirements for the accuracy of PCB defect detection in the actual production environment, so we propose two PCB defect detection frameworks with multiple model fusion including the defect detection by multi-model voting method (DDMV) and the defect detection by multi-model learning method (DDML). With the purpose of reducing wrong and missing detection, the DDMV and DDML integrate multiple defect detection networks with different fusion strategies. The effectiveness and accuracy of the proposed framework are verified with extensive experiments on two open-source PCB datasets. The experimental results demonstrate that the proposed DDMV and DDML are better than any other individual state-of-the-art PCB defect detection model in F1-score, and the area under curve value of DDML is also higher than that of any other individual detection model. Furthermore, compared with DDMV, the DDML with an automatic machine learning method achieves the best performance in PCB defect detection, and the F1-score on the two datasets can reach 99.7% and 95.6% respectively.印刷电路板(PCB)是电子产品不可或缺的组成部分,它决定了这些产品的质量。随着制造技术的发展和进步,PCB的布局和结构变得越来越复杂。然而,有效准确的PCB缺陷检测方法却很少。在实际生产环境中,对PCB缺陷检测的准确性有很高的要求,因此我们提出了两种包括多模型融合的PCB缺陷检测框架,包括多模型投票方法(DDMV)和多模型学习方法(DDML)的缺陷检测。为了减少错误和漏检,DDMV和DDML将多个具有不同融合策略的缺陷检测网络进行整合。通过对两个开源PCB数据集进行大量实验,验证了所提出框架的有效性和准确性。实验结果表明,所提出的DDMV和DDML在F1分数方面优于其他任何单独的最先进的PCB缺陷检测模型,而DDML的曲线下面积值也高于其他任何单独的检测模型。此外,与DDMV相比,使用自动机器学习方法的DDML在PCB缺陷检测方面达到了最好的性能,在两个数据集上的F1分数分别可达到99.7%和95.6%。
KW - printed circuit board (PCB)
KW - defect detection
KW - model fusion
KW - object detection model
UR - http://www.scopus.com/inward/record.url?scp=85134688025&partnerID=8YFLogxK
U2 - 10.1007/s12204-022-2471-0
DO - 10.1007/s12204-022-2471-0
M3 - Journal article
AN - SCOPUS:85134688025
SN - 1007-1172
VL - 28
SP - 717
EP - 727
JO - Journal of Shanghai Jiaotong University (Science)
JF - Journal of Shanghai Jiaotong University (Science)
IS - 6
ER -