A training and testing dataset
The dataset contains labelled images of thin-film modules collected
during the PEARL project.
This dataset is used in this paper:
@article{sovetkin2020encoder,
title={Encoder-decoder semantic segmentation models for electroluminescence images of thin-film modules},
author={Sovetkin, Evgenii and Weber, Thomas and Achterberg, Elbert Jan and Pieters, Bart Elger},
journal={IEEE Journal of Photovoltaics},
year={2020},
}
Structure of the data
The data is split onto 3 groups: training, testing and
final_evaluation. Each group contains labelled images for droplets and
shunts.
All images are located in "images" directory. All labelled images are
located in "labels" directory.
