ML Anthology

Hyperspectral Imaging and Computer Vision Based Remote Monitoring of SO2 Emissions in Maritime Vessels

Arnoud Jochemsen, Hege Indresand, Martin Chamberland, Etienne Drouin, Jan Robert Fiksdal, Xuan Zhang, Nabil Belbachir
ECCVW 2024 pp. 328-341
doi:10.1007/978-3-031-92805-5_21 /eccvw/2024/jochemsen2024eccvw-hyperspectral/

Abstract

This paper presents a hyperspectral camera system and data processing workflow designed to remotely detect, identify, and quantify $SO_2$ S O 2 emissions from ships in real-time, determining their fuel sulfur content (FSC). This technology is intended to assist maritime authorities in enforcement of the maritime sulfur emissions regulations. The focus of the study is on the automatic detection of ships and their exhaust plumes, enabling a fully automated verification of FSC. The system employs classic motion detection techniques, such as frame differencing and traditional computer vision morphological operations, to identify a ship, the plume and the chimney in a scene. A spectral angle mapper is the main method for finding segments in the hyperspectral data cubes. These simple methods can lead to a robust detection of the relevant scene pixels and calculation of the FSC from the spectra of these pixels.

PDF ECCVW Semantic Scholar

Cite

Text

Jochemsen et al. "Hyperspectral Imaging and Computer Vision Based Remote Monitoring of SO2 Emissions in Maritime Vessels." European Conference on Computer Vision Workshops, 2024. doi:10.1007/978-3-031-92805-5_21

Markdown

[Jochemsen et al. "Hyperspectral Imaging and Computer Vision Based Remote Monitoring of SO2 Emissions in Maritime Vessels." European Conference on Computer Vision Workshops, 2024.](https://mlanthology.org/eccvw/2024/jochemsen2024eccvw-hyperspectral/) doi:10.1007/978-3-031-92805-5_21

BibTeX

@inproceedings{jochemsen2024eccvw-hyperspectral,
  title     = {{Hyperspectral Imaging and Computer Vision Based Remote Monitoring of SO2 Emissions in Maritime Vessels}},
  author    = {Jochemsen, Arnoud and Indresand, Hege and Chamberland, Martin and Drouin, Etienne and Fiksdal, Jan Robert and Zhang, Xuan and Belbachir, Nabil},
  booktitle = {European Conference on Computer Vision Workshops},
  year      = {2024},
  pages     = {328-341},
  doi       = {10.1007/978-3-031-92805-5_21},
  url       = {https://mlanthology.org/eccvw/2024/jochemsen2024eccvw-hyperspectral/}
}