Convolution neural network based polycrystalline silicon photovoltaic cell linear defect diagnosis using electroluminescence images

作者：

Highlights：

• A three-phase algorithm is proposed for automatic linear defects diagnosis is proposed.

• The solution combined the advantages of the traditional image processing techique and deep learning.

• The solution obtain the best trade-off between computing accuracy and complexity.

• A dataset of PV module EL images is well established and maintained.

摘要

•A three-phase algorithm is proposed for automatic linear defects diagnosis is proposed.•The solution combined the advantages of the traditional image processing techique and deep learning.•The solution obtain the best trade-off between computing accuracy and complexity.•A dataset of PV module EL images is well established and maintained.

论文关键词：Electroluminescence images,Defects classification,Feature extraction,Deep learning

论文评审过程：Received 28 June 2021, Revised 21 September 2021, Accepted 28 March 2022, Available online 26 April 2022, Version of Record 27 April 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.117087