Premium
Deep learning‐based automatic detection of multitype defects in photovoltaic modules and application in real production line
Author(s) -
Zhao Yang,
Zhan Ke,
Wang Zhen,
Shen Wenzhong
Publication year - 2021
Publication title -
progress in photovoltaics: research and applications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.286
H-Index - 131
eISSN - 1099-159X
pISSN - 1062-7995
DOI - 10.1002/pip.3395
Subject(s) - confusion matrix , computer science , artificial intelligence , convolutional neural network , line (geometry) , production line , object detection , photovoltaic system , pattern recognition (psychology) , deep learning , compensation (psychology) , confusion , computer vision , engineering , mathematics , electrical engineering , mechanical engineering , psychology , geometry , psychoanalysis
Automatic defect detection in electroluminescence (EL) images of photovoltaic (PV) modules in production line remains as a challenge to replace time‐consuming and expensive human inspection and improve capacity. This paper presents a deep learning‐based automatic detection of multitype defects to fulfill inspection requirements of production line. At first, a database composed of 5983 labeled EL images of defective PV modules is built, and 19 types of identified defects are introduced. Next, a convolutional neural network is trained on top‐14 defects, and the best model is selected and tested, achieving 70.2% mAP 50 (mean average precision with at least 50% localization accuracy). Then, through analyzing an object detection‐based confusion matrix, recognition bias and detection compensation in missed defects that restrain the best model's mAP 50 are discovered to be harmless to normal/defective module classification in real production line. Finally, after setting specific screen criteria for different types of defects, normal/defective module classification is conducted on additionally collected 4791 EL images of PV modules on 3 days, and the best model achieves balanced scores of 95.1%, 96.0%, and 97.3%, respectively. As a result, this method surely has a highly promising potential to be adopted in real production line.