STTORM-CD low-demand and high-impact disaster monitoring
onboard satellites using change detection

J 2026

STTORM-CD low-demand and high-impact disaster monitoring onboard satellites using change detection

HEREC, Jonáš; Jan SEDMIDUBSKÝ a Rado PITOŇÁK

Základní údaje

Originální název

STTORM-CD low-demand and high-impact disaster monitoring onboard satellites using change detection

Autoři

HEREC, Jonáš; Jan SEDMIDUBSKÝ a Rado PITOŇÁK

Vydání

SCIENTIFIC REPORTS, Berlin, NATURE PORTFOLIO, 2026, 2045-2322

Další údaje

Jazyk

angličtina

Typ výsledku

Článek v odborném periodiku

Obor

10200 1.2 Computer and information sciences

Stát vydavatele

Velká Británie a Severní Irsko

Utajení

není předmětem státního či obchodního tajemství

Odkazy

URL

Impakt faktor

Impact factor: 3.900 v roce 2024

Označené pro přenos do RIV

Ano

Organizační jednotka

Fakulta informatiky

Klíčová slova anglicky

satellite imagery; flood detection; disaster management; change detection; Variational Autoencoder; Triplet loss; onboard processing

Štítky

DISA

Příznaky

Mezinárodní význam, Recenzováno

Změněno: 14. 4. 2026 14:05, doc. RNDr. Jan Sedmidubský, Ph.D.

Anotace

V originále

Satellite imagery can play a crucial role in disaster management, but critical images often take hours or even days to reach end-users, and upgrading hardware to improve transmission speed is prohibitively expensive for many small satellite missions. This article thus explores onboard change detection methods as a cost-effective alternative to reduce reaction time. Building on RaVAEn, we introduce STTORM-CD, a framework that combines a Variational Autoencoder (VAE) with a triplet loss, specifically designed for change detection. The triplet loss improves detection accuracy while maintaining the computational and storage efficiency of VAE, making it suitable for deployment on resource-constrained satellite hardware. To support training and evaluation, we present a new dataset, STTORM-CD-Floods, annotated with a custom strategy optimized for flood detection, along with new metrics, AURC and RDP, designed to address limitations of RaVAEn evaluation strategies, which are influenced more by dataset composition than model performance. Our experiments show that STTORM-CD outperforms existing flood detection methods, achieving an increase of 35 percentage points (pp) in custom AURC and standard AUPRC metrics against RaVAEn on the presented STTORM-CD-Floods dataset, while showing negligible changes in AURC (approximately -4 to +0.1 pp) for landslides and wildfires. This demonstrates that improvements on one disaster type do not necessarily compromise performance on others and highlights the potential for a universal and accurate real-time disaster detection system.

Návaznosti

MUNI/A/1873/2025, interní kód MU

Název: Komplexní data a systémy s využitím umělé inteligence

Investor: Masarykova univerzita, Komplexní data a systémy s využitím umělé inteligence

Citovat

HEREC, Jonáš; Jan SEDMIDUBSKÝ a Rado PITOŇÁK. STTORM-CD low-demand and high-impact disaster monitoring onboard satellites using change detection. SCIENTIFIC REPORTS. Berlin: NATURE PORTFOLIO, 2026, roč. 16, č. 1, s. 1-15. ISSN 2045-2322. Dostupné z: https://doi.org/10.1038/s41598-025-32598-3.

@article{2566320,
   author = {Herec, Jonáš and Sedmidubský, Jan and Pitoňák, Rado},
   article_location = {Berlin},
   article_number = {1},
   doi = {https://doi.org/10.1038/s41598-025-32598-3},
   keywords = {satellite imagery; flood detection; disaster management; change detection; Variational Autoencoder; Triplet loss; onboard processing},
   language = {eng},
   issn = {2045-2322},
   journal = {SCIENTIFIC REPORTS},
   title = {STTORM-CD low-demand and high-impact disaster monitoring onboard satellites using change detection},
   url = {https://www.nature.com/articles/s41598-025-32598-3},
   volume = {16},
   year = {2026}
}

TY  - JOUR
ID  - 2566320
AU  - Herec, Jonáš - Sedmidubský, Jan - Pitoňák, Rado
PY  - 2026
TI  - STTORM-CD low-demand and high-impact disaster monitoring onboard satellites using change detection
JF  - SCIENTIFIC REPORTS
VL  - 16
IS  - 1
SP  - 1-15
EP  - 1-15
PB  - NATURE PORTFOLIO
SN  - 20452322
KW  - satellite imagery
KW  - flood detection
KW  - disaster management
KW  - change detection
KW  - Variational Autoencoder
KW  - Triplet loss
KW  - onboard processing
UR  - https://www.nature.com/articles/s41598-025-32598-3
N2  - Satellite imagery can play a crucial role in disaster management, but critical images often take hours or even days to reach end-users, and upgrading hardware to improve transmission speed is prohibitively expensive for many small satellite missions. This article thus explores onboard change detection methods as a cost-effective alternative to reduce reaction time. Building on RaVAEn, we introduce STTORM-CD, a framework that combines a Variational Autoencoder (VAE) with a triplet loss, specifically designed for change detection. The triplet loss improves detection accuracy while maintaining the computational and storage efficiency of VAE, making it suitable for deployment on resource-constrained satellite hardware. To support training and evaluation, we present a new dataset, STTORM-CD-Floods, annotated with a custom strategy optimized for flood detection, along with new metrics, AURC and RDP, designed to address limitations of RaVAEn evaluation strategies, which are influenced more by dataset composition than model performance. Our experiments show that STTORM-CD outperforms existing flood detection methods, achieving an increase of 35 percentage points (pp) in custom AURC and standard AUPRC metrics against RaVAEn on the presented STTORM-CD-Floods dataset, while showing negligible changes in AURC (approximately -4 to +0.1 pp) for landslides and wildfires. This demonstrates that improvements on one disaster type do not necessarily compromise performance on others and highlights the potential for a universal and accurate real-time disaster detection system.
ER  -

HEREC, Jonáš; Jan SEDMIDUBSKÝ a Rado PITOŇÁK. STTORM-CD low-demand and high-impact disaster monitoring onboard satellites using change detection. \textit{SCIENTIFIC REPORTS}. Berlin: NATURE PORTFOLIO, 2026, roč.~16, č.~1, s.~1-15. ISSN~2045-2322. Dostupné z: https://doi.org/10.1038/s41598-025-32598-3.

Přehled o publikaci