README.md

---
license: cc-by-nc-4.0
pipeline_tag: image-segmentation
tags:
- remote sensing
- EMIT
- Hyperspectral
- AVIRIS
- methane
- CH4
---

# STARCOP trained models

This repository contains the trained models of the publication:

>  V. Růžička, G. Mateo-Garcia, L. Gómez-Chova, A. Vaughan, L. Guanter, and A. Markham. [Semantic segmentation of methane plumes with hyperspectral machine learning models](https://www.nature.com/articles/s41598-023-44918-6). _Scientific Reports 13, 19999_ (2023). DOI: 10.1038/s41598-023-44918-6.

We include the trained models:
* **HyperSTARCOP, only mag1c** in folder `models/hyperstarcop_mag1c_only`
* **HyperSTARCOP, mag1c + rgb**  in folder `models/hyperstarcop_mag1c_rgb`

The following tables shows the performance of the models in the AVIRIS test dataset and in the EMIT test dataset:
![metrics_starcop_aviris](table/table1.png)
![metrics_starcop_emit](table/table2.png)

In order to run any of these models see our tutorials in the [STARCOP repository](https://github.com/spaceml-org/STARCOP):

* [*STARCOP demo AVIRIS*](https://github.com/spaceml-org/STARCOP/blob/main/notebooks/model_demos_AVIRIS.ipynb).
* [*Run STARCOP models on raw EMIT data*](https://github.com/spaceml-org/STARCOP/blob/main/notebooks/inference_on_raw_EMIT_nc_file.ipynb).

If you find this work useful please cite:

```
@article{ruzicka_starcop_2023,
	title = {Semantic segmentation of methane plumes with hyperspectral machine learning models},
	volume = {13},
	issn = {2045-2322},
	url = {https://www.nature.com/articles/s41598-023-44918-6},
	doi = {10.1038/s41598-023-44918-6},
	number = {1},
	journal = {Scientific Reports},
	author = {Růžička, Vít and Mateo-Garcia, Gonzalo and Gómez-Chova, Luis and Vaughan, Anna, and Guanter, Luis and Markham, Andrew},
	month = nov,
	year = {2023},
	pages = {19999}
}
```