Model Adjusted Generalized Tests for Methane Plume Detection on Hyperspectral Images

E. OUERGHI; T. EHRET; G. FACCIOLO; E. MEINHARDT; J.-M. MOREL; C. DE FRANCHIS; T. LAUVAUX

doi:10.1109/WHISPERS61460.2023.10431079

Model Adjusted Generalized Tests for Methane Plume Detection on Hyperspectral Images

E. OUERGHI^*
, T. EHRET
, G. FACCIOLO
, E. MEINHARDT
, J.-M. MOREL
, C. DE FRANCHIS
, T. LAUVAUX

^*Corresponding author for this work

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Research › peer-review

1 Citation (Scopus)

Abstract

Reducing methane emissions is essential to tackle climate change. Here, we address the problem of detecting automatically point source methane leaks using high resolution hyperspectral images from the PRISMA satellite. We test the Generalized Likelihood Ratio Test (GLRT) for plume detection and compare it to the Matched Filter (MF). We then propose an improvement of the GLRT by using an adjustment coefficient. We introduce this new method under the name: Model Adjusted GLRT (MA-GLRT). We show that the MA-GLRT method reduces the fraction of false detections compared to the MF and the standard GLRT without preventing the detection of plumes. To validate the method, we use a dataset of manually annotated plumes on PRISMA images. We then show that our method outperforms the matched filter and the GLRT in terms of F1 score.

Original language	English
Title of host publication	Proceedings of the 13th Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing, WHISPERS 2023
Publisher	IEEE
ISBN (Electronic)	9798350395570
ISBN (Print)	9798350395587
DOIs	https://doi.org/10.1109/WHISPERS61460.2023.10431079
Publication status	Published - 2023
Externally published	Yes

Publication series

Name	Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing
ISSN (Print)	2158-6276

Bibliographical note

Publisher Copyright:
© 2023 IEEE.

Funding

Work partly financed by Office of Naval research grant N00014-20- S-B001, MENRT, and a PhD scholarship financed by MESRI (Minist`ere de l Enseignement Sup erieur, de la Recherche et de l Innovation). Centre Borelli is also a member of Universit e Paris Cit e, SSA and INSERM.

Keywords

Anomaly detection
Hyperspectral images
Manalahobis distance
Methane
Prisma

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/WHISPERS61460.2023.10431079

Cite this

OUERGHI, E., EHRET, T., FACCIOLO, G., MEINHARDT, E., MOREL, J.-M., DE FRANCHIS, C., & LAUVAUX, T. (2023). Model Adjusted Generalized Tests for Methane Plume Detection on Hyperspectral Images. In Proceedings of the 13th Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing, WHISPERS 2023 (Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing). IEEE. https://doi.org/10.1109/WHISPERS61460.2023.10431079

@inproceedings{586a6a45bb644d98a80a19cc3b306b5b,

title = "Model Adjusted Generalized Tests for Methane Plume Detection on Hyperspectral Images",

abstract = "Reducing methane emissions is essential to tackle climate change. Here, we address the problem of detecting automatically point source methane leaks using high resolution hyperspectral images from the PRISMA satellite. We test the Generalized Likelihood Ratio Test (GLRT) for plume detection and compare it to the Matched Filter (MF). We then propose an improvement of the GLRT by using an adjustment coefficient. We introduce this new method under the name: Model Adjusted GLRT (MA-GLRT). We show that the MA-GLRT method reduces the fraction of false detections compared to the MF and the standard GLRT without preventing the detection of plumes. To validate the method, we use a dataset of manually annotated plumes on PRISMA images. We then show that our method outperforms the matched filter and the GLRT in terms of F1 score.",

keywords = "Anomaly detection, Hyperspectral images, Manalahobis distance, Methane, Prisma",

author = "E. OUERGHI and T. EHRET and G. FACCIOLO and E. MEINHARDT and J.-M. MOREL and \{DE FRANCHIS\}, C. and T. LAUVAUX",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.",

year = "2023",

doi = "10.1109/WHISPERS61460.2023.10431079",

language = "English",

isbn = "9798350395587",

series = "Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing",

publisher = "IEEE",

booktitle = "Proceedings of the 13th Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing, WHISPERS 2023",

}

OUERGHI, E, EHRET, T, FACCIOLO, G, MEINHARDT, E, MOREL, J-M, DE FRANCHIS, C & LAUVAUX, T 2023, Model Adjusted Generalized Tests for Methane Plume Detection on Hyperspectral Images. in Proceedings of the 13th Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing, WHISPERS 2023. Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing, IEEE. https://doi.org/10.1109/WHISPERS61460.2023.10431079

Model Adjusted Generalized Tests for Methane Plume Detection on Hyperspectral Images. / OUERGHI, E.; EHRET, T.; FACCIOLO, G. et al.
Proceedings of the 13th Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing, WHISPERS 2023. IEEE, 2023. (Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing).

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Research › peer-review

TY - GEN

T1 - Model Adjusted Generalized Tests for Methane Plume Detection on Hyperspectral Images

AU - OUERGHI, E.

AU - EHRET, T.

AU - FACCIOLO, G.

AU - MEINHARDT, E.

AU - MOREL, J.-M.

AU - DE FRANCHIS, C.

AU - LAUVAUX, T.

PY - 2023

Y1 - 2023

N2 - Reducing methane emissions is essential to tackle climate change. Here, we address the problem of detecting automatically point source methane leaks using high resolution hyperspectral images from the PRISMA satellite. We test the Generalized Likelihood Ratio Test (GLRT) for plume detection and compare it to the Matched Filter (MF). We then propose an improvement of the GLRT by using an adjustment coefficient. We introduce this new method under the name: Model Adjusted GLRT (MA-GLRT). We show that the MA-GLRT method reduces the fraction of false detections compared to the MF and the standard GLRT without preventing the detection of plumes. To validate the method, we use a dataset of manually annotated plumes on PRISMA images. We then show that our method outperforms the matched filter and the GLRT in terms of F1 score.

AB - Reducing methane emissions is essential to tackle climate change. Here, we address the problem of detecting automatically point source methane leaks using high resolution hyperspectral images from the PRISMA satellite. We test the Generalized Likelihood Ratio Test (GLRT) for plume detection and compare it to the Matched Filter (MF). We then propose an improvement of the GLRT by using an adjustment coefficient. We introduce this new method under the name: Model Adjusted GLRT (MA-GLRT). We show that the MA-GLRT method reduces the fraction of false detections compared to the MF and the standard GLRT without preventing the detection of plumes. To validate the method, we use a dataset of manually annotated plumes on PRISMA images. We then show that our method outperforms the matched filter and the GLRT in terms of F1 score.

KW - Anomaly detection

KW - Hyperspectral images

KW - Manalahobis distance

KW - Methane

KW - Prisma

UR - https://www.scopus.com/pages/publications/85186266893

U2 - 10.1109/WHISPERS61460.2023.10431079

DO - 10.1109/WHISPERS61460.2023.10431079

M3 - Conference paper (refereed)

AN - SCOPUS:85186266893

SN - 9798350395587

T3 - Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing

BT - Proceedings of the 13th Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing, WHISPERS 2023

PB - IEEE

ER -

OUERGHI E, EHRET T, FACCIOLO G, MEINHARDT E, MOREL JM, DE FRANCHIS C et al. Model Adjusted Generalized Tests for Methane Plume Detection on Hyperspectral Images. In Proceedings of the 13th Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing, WHISPERS 2023. IEEE. 2023. (Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing). doi: 10.1109/WHISPERS61460.2023.10431079

Model Adjusted Generalized Tests for Methane Plume Detection on Hyperspectral Images

Abstract

Publication series

Bibliographical note

Funding

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this