ModelPublicadoExterno

Fast Methane Filters Models

This Hugging Face repository contains models trained in the article "Optimizing Methane Detection On Board Satellites: Speed, Accuracy, and Low-Power Solutions for Resource-Constrained Hardware."

Fuentehuggingface.co/onboard-coop/fast-methane-filters-models/tree/3a562945f128de46dbaeb2282323c4371b1dbb86

TipoOtherVersiónFinal checkpoint model 40epCcalidad

Fuente del modelo

Tipo: Other
Tamaño del archivo: 3,34–76,3 MB
Versión: vFinal checkpoint model 40ep
Actualizado: 1 nov 2026
Fuente: Hugging Face

Descripción de la fuente

Model Card

This Hugging Face repository contains models trained in the article "Optimizing Methane Detection On Board Satellites: Speed, Accuracy, and Low-Power Solutions for Resource-Constrained Hardware."

Paper: Optimizing Methane Detection On Board Satellites: Speed, Accuracy, and Low-Power Solutions for Resource-Constrained Hardware

Model Overview

The models here were trained using the code available at the following GitHub repository:

Training Code: HyperspectralViTs

The main project code, including filters benchmarking and demos, is available at the:

Project Code: Methane Filters Benchmark

Data and Products

The precomputed products used for training were created by code in the main project repository:

Products Creation Code: methane-filters-benchmark

Additionally, these precomputed products are hosted and accessible here:

Dataset Repository: STARCOP-fast-products

Sample Usage

You can try out our models and demos directly in Google Colab using the provided notebooks:

Models Demo: Open In Colab This notebook demonstrates model inference.
Products Creation and Benchmarking Demo: Open In Colab This notebook demonstrates generating products and measuring their runtime.

For local inference using the ONNX models, refer to the benchmark/onnx_inference_time.py script in the Project Code repository.

Citation

If you use these models in your research, please cite our article:

@misc{herec2026fastmethanedetectionpipeline,
      title={A Fast Methane Detection Pipeline on Board Satellites Based on Mag1c-SAS and LinkNet}, 
      author={Jonáš Herec and Vít Růžička and Rado Pitoňák and Jan Sedmidubsky},
      year={2026},
      eprint={2606.03675},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2606.03675}, 
}

Models performance

U-Net - CEM

ID	Recall	Precision	F1	F1 strong
A	0.441	0.317	0.369	0.500
B	0.701	0.158	0.258	0.550
C	0.531	0.299	0.382	0.610
D	0.536	0.218	0.310	0.551
E	0.564	0.182	0.275	0.469
AVG	55.47%	23.49%	31.90%	53.55%
STD	8.41%	6.30%	4.94%	4.85%

U-Net - ACE

ID	Recall	Precision	F1	F1 strong
A	0.468	0.202	0.282	0.460
B	0.480	0.288	0.360	0.537
C	0.413	0.253	0.314	0.461
D	0.550	0.194	0.287	0.510
E	0.500	0.162	0.245	0.442
AVG	48.22%	21.99%	29.77%	48.19%
STD	4.46%	4.50%	3.82%	3.57%

U-Net - MF

ID	Recall	Precision	F1	F1 strong
A	0.603	0.153	0.243	0.451
B	0.673	0.198	0.306	0.585
C	0.563	0.259	0.355	0.507
D	0.625	0.173	0.271	0.558
E	0.466	0.301	0.366	0.496
AVG	58.60%	21.68%	30.82%	51.94%
STD	6.97%	5.52%	4.73%	4.73%

U-Net - MAG1C-SAS

ID	Recall	Precision	F1	F1 strong
A	0.587	0.456	0.513	0.668
B	0.618	0.291	0.395	0.642
C	0.576	0.290	0.386	0.604
D	0.613	0.414	0.495	0.686
E	0.427	0.280	0.338	0.470
AVG	56.42%	34.62%	42.54%	61.40%
STD	7.04%	7.38%	6.73%	7.71%

U-Net - MAG1C (tile-wise)

ID	Recall	Precision	F1	F1 strong
A	0.643	0.218	0.325	0.599
B	0.732	0.288	0.413	0.692
C	0.613	0.362	0.455	0.659
D	0.669	0.242	0.355	0.633
E	0.640	0.366	0.466	0.684
AVG	65.94%	29.52%	40.28%	65.34%
STD	4.04%	6.05%	5.51%	3.42%

LinkNet - CEM

ID	Recall	Precision	F1	F1 strong
A	0.597	0.319	0.416	0.633
B	0.539	0.274	0.363	0.603
C	0.452	0.233	0.308	0.527
D	0.606	0.165	0.260	0.561
E	0.442	0.144	0.217	0.455
AVG	52.72%	22.70%	31.27%	55.56%
STD	6.96%	6.54%	7.09%	6.20%

LinkNet - MAG1C-SAS

ID	Recall	Precision	F1	F1 strong
A	0.566	0.324	0.412	0.612
B	0.505	0.515	0.510	0.613
C	0.381	0.422	0.400	0.507
D	0.590	0.383	0.464	0.660
E	0.513	0.378	0.435	0.627
AVG	51.10%	40.44%	44.42%	60.38%
STD	7.24%	6.35%	3.95%	5.14%

U-Net - ACE_multiplicative

ID	Recall	Precision	F1	F1 strong
A	0.458	0.382	0.417	0.510
B	0.391	0.374	0.382	0.472
C	0.600	0.262	0.365	0.593
D	0.409	0.344	0.374	0.449
E	0.516	0.324	0.398	0.542
AVG	47.50%	33.72%	38.72%	51.32%
STD	7.60%	4.30%	1.84%	5.09%

U-Net - MF_multiplicative

ID	Recall	Precision	F1	F1 strong
A	0.452	0.468	0.459	0.574
B	0.638	0.451	0.528	0.707
C	0.557	0.311	0.399	0.566
D	0.518	0.322	0.397	0.582
E	0.705	0.251	0.370	0.637
AVG	57.41%	36.05%	43.09%	61.32%
STD	8.92%	8.43%	5.68%	5.31%

Fuentes

1 fuente

Hugging FacePrincipalVersión 3a562945f128de46dbaeb2282323c4371b1dbb86

huggingface.co/onboard-coop/fast-methane-filters-models/tree/3a562945f128de46dbaeb2282323c4371b1dbb86

Verificado 1 nov

Download files

Artefactos del modelo

64 artefactos

LINKNET/CEM/LinkNet_RGB_CEM_A/final_checkpoint_model_40ep.ckpt

ckpt · 10,1 MB · SHA-256 a9ed4d07a14c…b38a · Hugging Face

Download file

LINKNET/CEM/LinkNet_RGB_CEM_A/final_checkpoint_model_40ep.onnx

onnx · 3,34 MB · SHA-256 983240a72f36…01ae · Hugging Face

Download file

LINKNET/CEM/LinkNet_RGB_CEM_B/final_checkpoint_model_40ep.ckpt

ckpt · 10,1 MB · SHA-256 fe61719b41c1…2a7f · Hugging Face

Download file

LINKNET/CEM/LinkNet_RGB_CEM_B/final_checkpoint_model_40ep.onnx

onnx · 3,34 MB · SHA-256 4453a12325e5…4736 · Hugging Face

Download file

LINKNET/CEM/LinkNet_RGB_CEM_C/final_checkpoint_model_40ep.ckpt

ckpt · 10,1 MB · SHA-256 54160ce86fc5…cd21 · Hugging Face

Download file

LINKNET/CEM/LinkNet_RGB_CEM_C/final_checkpoint_model_40ep.onnx

onnx · 3,34 MB · SHA-256 2c2aead24299…8e76 · Hugging Face

Download file

LINKNET/CEM/LinkNet_RGB_CEM_D/final_checkpoint_model_40ep.ckpt

ckpt · 10,1 MB · SHA-256 3e170bf1d3c5…7daa · Hugging Face

Download file

LINKNET/CEM/LinkNet_RGB_CEM_D/final_checkpoint_model_40ep.onnx

onnx · 3,34 MB · SHA-256 3249e151404c…ede5 · Hugging Face

Download file

LINKNET/CEM/LinkNet_RGB_CEM_E/final_checkpoint_model_40ep.ckpt

ckpt · 10,1 MB · SHA-256 ae561622c2bb…4262 · Hugging Face

Download file

LINKNET/CEM/LinkNet_RGB_CEM_E/final_checkpoint_model_40ep.onnx

onnx · 3,34 MB · SHA-256 2041a7527090…8f9b · Hugging Face

Download file

LINKNET/Mag1c-SAS/LinkNet_RGB_Mag1cSAS_A/final_checkpoint_model_40ep.ckpt

ckpt · 10,1 MB · SHA-256 bc093ddc5d68…0c65 · Hugging Face

Download file

LINKNET/Mag1c-SAS/LinkNet_RGB_Mag1cSAS_A/final_checkpoint_model_40ep.onnx

onnx · 3,34 MB · SHA-256 a85658456004…3203 · Hugging Face

Download file

LINKNET/Mag1c-SAS/LinkNet_RGB_Mag1cSAS_B/final_checkpoint_model_40ep.ckpt

ckpt · 10,1 MB · SHA-256 672448f51260…c409 · Hugging Face

Download file

LINKNET/Mag1c-SAS/LinkNet_RGB_Mag1cSAS_B/final_checkpoint_model_40ep.onnx

onnx · 3,34 MB · SHA-256 b36f7f24c36a…44f3 · Hugging Face

Download file

LINKNET/Mag1c-SAS/LinkNet_RGB_Mag1cSAS_C/final_checkpoint_model_40ep.ckpt

ckpt · 10,1 MB · SHA-256 4fd1e36a27da…88c6 · Hugging Face

Download file

LINKNET/Mag1c-SAS/LinkNet_RGB_Mag1cSAS_C/final_checkpoint_model_40ep.onnx

onnx · 3,34 MB · SHA-256 966625fcdb21…65f8 · Hugging Face

Download file

LINKNET/Mag1c-SAS/LinkNet_RGB_Mag1cSAS_D/final_checkpoint_model_40ep.ckpt

ckpt · 10,1 MB · SHA-256 f2f8234c522d…b884 · Hugging Face

Download file

LINKNET/Mag1c-SAS/LinkNet_RGB_Mag1cSAS_D/final_checkpoint_model_40ep.onnx

onnx · 3,34 MB · SHA-256 b2943627b6d2…3cb1 · Hugging Face

Download file

LINKNET/Mag1c-SAS/LinkNet_RGB_Mag1cSAS_E/final_checkpoint_model_40ep.ckpt

ckpt · 10,1 MB · SHA-256 9e6f56b5450d…0746 · Hugging Face

Download file

LINKNET/Mag1c-SAS/LinkNet_RGB_Mag1cSAS_E/final_checkpoint_model_40ep.onnx

onnx · 3,34 MB · SHA-256 4de1c92fbf3c…663c · Hugging Face

Download file

UNET/ACE_multiplicative/UNet_RGB_ACE_multiplicative_A/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 a8db4c002d8f…56a1 · Hugging Face

Download file

UNET/ACE_multiplicative/UNet_RGB_ACE_multiplicative_B/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 e327534a9438…16a4 · Hugging Face

Download file

UNET/ACE_multiplicative/UNet_RGB_ACE_multiplicative_C/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 9ea65f61ab08…2794 · Hugging Face

Download file

UNET/ACE_multiplicative/UNet_RGB_ACE_multiplicative_D/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 a905dd404965…7b9a · Hugging Face

Download file

UNET/ACE_multiplicative/UNet_RGB_ACE_multiplicative_E/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 8e1e8452a118…4750 · Hugging Face

Download file

UNET/ACE/UNet_RGB_ACE_A/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 6428a39ee079…8bb1 · Hugging Face

Download file

UNET/ACE/UNet_RGB_ACE_A/final_checkpoint_model_40ep.onnx

onnx · 25,3 MB · SHA-256 fa3b2ecb6b62…f558 · Hugging Face

Download file

UNET/ACE/UNet_RGB_ACE_B/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 33fa50ea7ebf…f748 · Hugging Face

Download file

UNET/ACE/UNet_RGB_ACE_B/final_checkpoint_model_40ep.onnx

onnx · 25,3 MB · SHA-256 4e9933d60f68…91e9 · Hugging Face

Download file

UNET/ACE/UNet_RGB_ACE_C/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 3a845006d70b…0a9d · Hugging Face

Download file

UNET/ACE/UNet_RGB_ACE_C/final_checkpoint_model_40ep.onnx

onnx · 25,3 MB · SHA-256 ff7411d49089…bd4e · Hugging Face

Download file

UNET/ACE/UNet_RGB_ACE_D/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 ba94b0fafbc5…22fb · Hugging Face

Download file

UNET/ACE/UNet_RGB_ACE_D/final_checkpoint_model_40ep.onnx

onnx · 25,3 MB · SHA-256 ef7e6985ab32…6cd4 · Hugging Face

Download file

UNET/ACE/UNet_RGB_ACE_E/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 9799a05cceb2…4743 · Hugging Face

Download file

UNET/ACE/UNet_RGB_ACE_E/final_checkpoint_model_40ep.onnx

onnx · 25,3 MB · SHA-256 1617e4c98c7d…cc89 · Hugging Face

Download file

UNET/CEM/UNet_RGB_CEM_A/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 ae047c5d6003…f3c4 · Hugging Face

Download file

UNET/CEM/UNet_RGB_CEM_A/final_checkpoint_model_40ep.onnx

onnx · 25,3 MB · SHA-256 518c13a5431d…ddcf · Hugging Face

Download file

UNET/CEM/UNet_RGB_CEM_B/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 9bf6e389835f…3e02 · Hugging Face

Download file

UNET/CEM/UNet_RGB_CEM_B/final_checkpoint_model_40ep.onnx

onnx · 25,3 MB · SHA-256 e00e801f18f1…4d0b · Hugging Face

Download file

UNET/CEM/UNet_RGB_CEM_C/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 a54fada64ac8…3c39 · Hugging Face

Download file

UNET/CEM/UNet_RGB_CEM_C/final_checkpoint_model_40ep.onnx

onnx · 25,3 MB · SHA-256 263e7dac5593…e19d · Hugging Face

Download file

UNET/CEM/UNet_RGB_CEM_D/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 8d455164e892…9926 · Hugging Face

Download file

UNET/CEM/UNet_RGB_CEM_D/final_checkpoint_model_40ep.onnx

onnx · 25,3 MB · SHA-256 06312d3feed7…68b8 · Hugging Face

Download file

UNET/CEM/UNet_RGB_CEM_E/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 9c8ea1f68325…1a6b · Hugging Face

Download file

UNET/CEM/UNet_RGB_CEM_E/final_checkpoint_model_40ep.onnx

onnx · 25,3 MB · SHA-256 6df1ab5c9354…2ff7 · Hugging Face

Download file

UNET/Mag1c-SAS/UNet_RGB_Mag1cSAS_A/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 45a42183218d…0f61 · Hugging Face

Download file

UNET/Mag1c-SAS/UNet_RGB_Mag1cSAS_A/final_checkpoint_model_40ep.onnx

onnx · 25,3 MB · SHA-256 65c5b48e0ea4…c491 · Hugging Face

Download file

UNET/Mag1c-SAS/UNet_RGB_Mag1cSAS_B/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 a8b631768656…997e · Hugging Face

Download file

UNET/Mag1c-SAS/UNet_RGB_Mag1cSAS_B/final_checkpoint_model_40ep.onnx

onnx · 25,3 MB · SHA-256 3318914db17f…233d · Hugging Face

Download file

UNET/Mag1c-SAS/UNet_RGB_Mag1cSAS_C/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 dde296053253…ff60 · Hugging Face

Download file

UNET/Mag1c-SAS/UNet_RGB_Mag1cSAS_C/final_checkpoint_model_40ep.onnx

onnx · 25,3 MB · SHA-256 8b044731bbfa…efed · Hugging Face

Download file

UNET/Mag1c-SAS/UNet_RGB_Mag1cSAS_D/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 52dc351eeb2a…7a4c · Hugging Face

Download file

UNET/Mag1c-SAS/UNet_RGB_Mag1cSAS_D/final_checkpoint_model_40ep.onnx

onnx · 25,3 MB · SHA-256 b882dce326f1…f916 · Hugging Face

Download file

UNET/Mag1c-SAS/UNet_RGB_Mag1cSAS_E/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 e163ee297cbf…965e · Hugging Face

Download file

UNET/Mag1c-SAS/UNet_RGB_Mag1cSAS_E/final_checkpoint_model_40ep.onnx

onnx · 25,3 MB · SHA-256 3452cfeb1da7…1708 · Hugging Face

Download file

UNET/Mag1c-tile/UNet_RGB_tileMag1c_A/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 7a53588cffc8…c774 · Hugging Face

Download file

UNET/Mag1c-tile/UNet_RGB_tileMag1c_A/final_checkpoint_model_40ep.onnx

onnx · 25,3 MB · SHA-256 8fe01080b97c…9d23 · Hugging Face

Download file

UNET/Mag1c-tile/UNet_RGB_tileMag1c_B/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 ed7967620b36…435d · Hugging Face

Download file

UNET/Mag1c-tile/UNet_RGB_tileMag1c_B/final_checkpoint_model_40ep.onnx

onnx · 25,3 MB · SHA-256 b7461233c509…ecf9 · Hugging Face

Download file

UNET/Mag1c-tile/UNet_RGB_tileMag1c_C/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 5748f36039e3…993c · Hugging Face

Download file

UNET/Mag1c-tile/UNet_RGB_tileMag1c_C/final_checkpoint_model_40ep.onnx

onnx · 25,3 MB · SHA-256 5d73193b59c1…0691 · Hugging Face

Download file

UNET/Mag1c-tile/UNet_RGB_tileMag1c_D/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 8b72cc11a04e…a4d1 · Hugging Face

Download file

UNET/Mag1c-tile/UNet_RGB_tileMag1c_D/final_checkpoint_model_40ep.onnx

onnx · 25,3 MB · SHA-256 88631f8d18fb…aa1d · Hugging Face

Download file

UNET/Mag1c-tile/UNet_RGB_tileMag1c_E/final_checkpoint_model_40ep.ckpt

ckpt · 76,3 MB · SHA-256 93c8ecbab0f7…3e7c · Hugging Face

Download file

Extractos de fuentes

2 extractos

Readme18 agohuggingface.co/onboard-coop/fast-methane-filters-models

--- license: cc-by-4.0 pipeline_tag: image-segmentation --- # Model Card This Hugging Face repository contains models trained in the article **"Optimizing Methane Detection On Board Satellites: Speed, Accuracy, and Low-Power Solutions for Resource-Constrained Hardware."** **Paper:** [Optimizing Methane Detection On Board Satellites: Speed, Accuracy, and Low-Power Solutions for Resource-Constrained Hardware](https://huggingface.co/papers/2507.01472) ## Model Overview The models here were trained using the code available at the following GitHub repository: * **Training Code**: [HyperspectralViTs](https://github.com/previtus/HyperspectralViTs) The main project code, including filters benchmarking and demos, is available at the: * **Project Code**: [Methane Filters Benchmark](https://github.com/zaitra/methane-filters-benchmark) ## Data and Products The precomputed products used for training were created by code in the main project repository: * **Products Creation Code**: [methane-filters-benchmark](https://github.com/zaitra/methane-filters-benchmark) Additionally, these precomputed products are hosted and accessible here: * **Dataset Repository**: [STARCOP-fast-products](https://huggingface.co/datasets/onboard-coop/STARCOP-fast-products) ## Sample Usage You can try out our models and demos directly in Google Colab using the provided notebooks: * **Models Demo**: [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/zaitra/methane-filters-benchmark/blob/main/ntbs/Models_demo.ipynb) This notebook demonstrates model inference. * **Products Creation and Benchmarking Demo**: [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/zaitra/methane-filters-benchmark/blob/main/ntbs/Products_demo.ipynb) This notebook demonstrates generating products and measuring their runtime. For local inference using the ONNX models, refer to the `benchmark/onnx_inference_time.py` script in the [Project Code repository](https://github.com/zaitra/methane-filters-benchmark). ## Citation If you use these models in your research, please cite our article: ```bibtex @misc{herec2026fastmethanedetectionpipeline, title={A Fast Methane Detection Pipeline on Board Satellites Based on Mag1c-SAS and LinkNet}, author={Jonáš Herec and Vít Růžička and Rad...

Note18 agohuggingface.co/onboard-coop/fast-methane-filters-models

Source context: 0 downloads · 1 likes · Pipeline image-segmentation · Repo onboard-coop/fast-methane-filters-models