Buerhop-Lutz C, Deitsch S, Maier A, Gallwitz F, Berger S, Doll B, Hauch J, Camus C, Brabec C (2018)
Publication Language: English
Publication Type: Conference contribution, Conference Contribution
Publication year: 2018
Pages Range: 1287-1289
ISBN: 3-936338-50-7
DOI: 10.4229/35thEUPVSEC20182018-5CV.3.15
In this paper a dataset consisting of 2,426 solar cells extracted from high-resolution electroluminescence (EL) images is used for automated defect probability recognition. The collected images contain both functional and defective solar cells with varying degrees of degradation both in monocrystalline and polycrystalline solar modules. The images were labeled by expert who categorized the solar cells by the likelihood of a defect within each image. The labeled images can be used for development of computer vision and machine learning methods for automatic detection of different defects, like cracks, fracture interconnects, PID, and cell quality and for the purpose of predicting the power efficiency loss.
APA:
Buerhop-Lutz, C., Deitsch, S., Maier, A., Gallwitz, F., Berger, S., Doll, B.,... Brabec, C. (2018). A Benchmark for Visual Identification of Defective Solar Cells in Electroluminescence Imagery. In Proceedings of the 35th European Photovoltaic Solar Energy Conference and Exhibition (pp. 1287-1289). Brussels, BE.
MLA:
Buerhop-Lutz, Claudia, et al. "A Benchmark for Visual Identification of Defective Solar Cells in Electroluminescence Imagery." Proceedings of the 35th European Photovoltaic Solar Energy Conference and Exhibition, Brussels 2018. 1287-1289.
BibTeX: Download