Update README.md
Browse files
README.md
CHANGED
|
@@ -1,4 +1,48 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
pipeline_tag: image-classification
|
| 4 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
pipeline_tag: image-classification
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
# Luminance Transformer
|
| 7 |
+
|
| 8 |
+
This repository contains code for Solar Transformer for electroluminescence (EL) images. The model is based on the transformer architecture and is designed to process EL images of solar cells.
|
| 9 |
+
|
| 10 |
+
## Background
|
| 11 |
+
EL imaging is a technique used to study solar cells. It involves capturing images of solar cells using a camera sensitive to the near-infrared region of the electromagnetic spectrum. These images show the distribution of charge carriers in the solar cell, which is related to the efficiency of the cell.
|
| 12 |
+
|
| 13 |
+
The solar-transformer model is designed to process EL images and predict the efficiency of the solar cell. It is based on the transformer architecture, which has been shown to be effective for processing sequential data such as natural language text.
|
| 14 |
+
|
| 15 |
+
## Results
|
| 16 |
+
|
| 17 |
+
The solar-transformer model achieves state-of-the-art performance on the EL image dataset, with an accuracy of 91.7% on a classfication test. (defective or functional)
|
| 18 |
+
|
| 19 |
+
ELPV-Monocystalline
|
| 20 |
+
| Model | Recall | Precision | F1-Score |
|
| 21 |
+
|---|---|---|---|
|
| 22 |
+
| Lumi-T | 0.9191 | 0.9339 | 0.9256 |
|
| 23 |
+
| VGG-19 | 0.8529 | 0.8603 | 0.8492 |
|
| 24 |
+
| ResNet-50 | 0.8824 | 0.8855 | 0.8806 |
|
| 25 |
+
|
| 26 |
+
ELPV-Polycystalline
|
| 27 |
+
| Model | Recall | Precision | F1-Score |
|
| 28 |
+
|---|---|---|---|
|
| 29 |
+
| Lumi-T | 0.9116 | 0.9509 | 0.9289 |
|
| 30 |
+
| VGG-19 | 0.8462 | 0.8729 | 0.8462 |
|
| 31 |
+
| ResNet-50 | 0.8269 | 0.8601 | 0.7951 |
|
| 32 |
+
ELPV-Overall
|
| 33 |
+
| Model | Recall | Precision | F1-Score |
|
| 34 |
+
|---|---|---|---|
|
| 35 |
+
| Lumi-T | 0.8851 | 0.9278 | 0.9170 |
|
| 36 |
+
| VGG-19 | 0.8552 | 0.8552 | 0.7885 |
|
| 37 |
+
| ResNet-50 | 0.8049 | 0.8476 | 0.8049 |
|
| 38 |
+
|
| 39 |
+
|
| 40 |
+
ELPV-Transfer Learning (Monocrystalline to Polycrystalline)
|
| 41 |
+
| Model | F1-Score |
|
| 42 |
+
|---|---|
|
| 43 |
+
| Lumi-T | 0.8202 |
|
| 44 |
+
| ResNet-50 | 0.6103 |
|
| 45 |
+
|
| 46 |
+
## Acknowledgments
|
| 47 |
+
|
| 48 |
+
This work was supported by the University of New South Wales KATANA HPC. We would also like to thank the support of GreenDyanmics Pty. Ltd.
|