Automatic Classification of Defective Photovoltaic Module Cells in Electroluminescence Images
Published:
- Who: the group project (3 people including myself)
- I was in charge of preliminary CNN test section and result analysis.
- Other 2 people were in charge of machine learning and transfer learning section, respectively.
- When: Mar. 2020 - June 2020
- Where: ECE228 (Machine Learning for Physical Applications) group project
- What: Classification of defective PV cells in Electroluminescence images (@ 6-inch, industrial level)
- Dataset: Open source in Github
- How: Python. OpenCV. TensorFlow 2.x. Keras data augmentation.
- Learn: TensorFlow 2.x usage. Image data augmentation libraries. Building vanilla CNN models from the scratch. Transfer learning. KAZE descriptor.
- The repository is now opened for the public.
- It will be back to the private in Aug. 2021. Please contact me if you want to see in detail.
- Featureed slides (Not full materials, only what I contributed to)
