SHAZAM: Self-Supervised Change Monitoring for Hazard Detection and Mapping

The increasing frequency of hazardous events due to climate change underscores an urgent need for effective hazard monitoring systems. Current approaches rely on expensive labelled datasets, struggle to distinguish hazardous changes from seasonal changes, or require multiple observations for to conf...

Full description

Saved in:
Bibliographic Details
Main Authors: Samuel Garske, Konrad Heidler, Bradley Evans, Kee Choon Wong, Xiao Xiang Zhu
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11033198/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850109371304902656
author Samuel Garske
Konrad Heidler
Bradley Evans
Kee Choon Wong
Xiao Xiang Zhu
author_facet Samuel Garske
Konrad Heidler
Bradley Evans
Kee Choon Wong
Xiao Xiang Zhu
author_sort Samuel Garske
collection DOAJ
description The increasing frequency of hazardous events due to climate change underscores an urgent need for effective hazard monitoring systems. Current approaches rely on expensive labelled datasets, struggle to distinguish hazardous changes from seasonal changes, or require multiple observations for to confirm when a potential hazardous event has occurred. To address these challenges, this work presents Self-Supervised Change Monitoring for Hazard Detection and Mapping—SHAZAM. SHAZAM uses a lightweight conditional UNet model to generate images of a region of interest (ROI) for any day of the year, allowing for the direct modelling of seasonal changes. A modified structural similarity measure compares the generated images with real satellite observations to compute both image-level anomaly scores, and pixel-level hazard maps. In addition, a seasonal threshold is introduced to further reduce the need for dataset-specific optimisation. SHAZAM was evaluated on four diverse datasets that contain wildfires, burned regions, extreme and out-of-season snowfall, floods, droughts, algal blooms, and deforestation, and achieved F1 score improvements of between 0.066 and 0.234 over existing methods. This was achieved primarily through a higher recall over other models, while using only 473 K parameters. SHAZAM demonstrates superior mapping capabilities through a higher spatial resolution and the suppression of background features for both immediate and gradual hazards. SHAZAM offers an effective and generalisable solution for hazard detection and mapping across different geographical regions and a diverse range of hazards, and highlights the potential of further self-supervised method contributions for hazard detection and mapping.
format Article
id doaj-art-a78f662f33a34f759fd1fa7e6c467b4d
institution OA Journals
issn 1939-1404
2151-1535
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
spelling doaj-art-a78f662f33a34f759fd1fa7e6c467b4d2025-08-20T02:38:05ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118154561547610.1109/JSTARS.2025.357933011033198SHAZAM: Self-Supervised Change Monitoring for Hazard Detection and MappingSamuel Garske0https://orcid.org/0000-0002-0622-8255Konrad Heidler1https://orcid.org/0000-0001-8226-0727Bradley Evans2https://orcid.org/0000-0001-6675-3118Kee Choon Wong3https://orcid.org/0000-0002-4977-5611Xiao Xiang Zhu4https://orcid.org/0000-0001-8107-9096School of Aerospace, Mechanical, and Mechatronic Engineering, The University of Sydney, Sydney, NSW, AustraliaSiPEO, Technical University of Munich, Munich, GermanySchool of Environmental and Rural Science, The University of New England, Armidale, NSW, AustraliaSchool of Aerospace, Mechanical, and Mechatronic Engineering, The University of Sydney, Sydney, NSW, AustraliaChair of Data Science in Earth Observation (SiPEO), Department of Aerospace and Geodesy, School of Engineering and Design, Technical University of Munich, Munich, GermanyThe increasing frequency of hazardous events due to climate change underscores an urgent need for effective hazard monitoring systems. Current approaches rely on expensive labelled datasets, struggle to distinguish hazardous changes from seasonal changes, or require multiple observations for to confirm when a potential hazardous event has occurred. To address these challenges, this work presents Self-Supervised Change Monitoring for Hazard Detection and Mapping—SHAZAM. SHAZAM uses a lightweight conditional UNet model to generate images of a region of interest (ROI) for any day of the year, allowing for the direct modelling of seasonal changes. A modified structural similarity measure compares the generated images with real satellite observations to compute both image-level anomaly scores, and pixel-level hazard maps. In addition, a seasonal threshold is introduced to further reduce the need for dataset-specific optimisation. SHAZAM was evaluated on four diverse datasets that contain wildfires, burned regions, extreme and out-of-season snowfall, floods, droughts, algal blooms, and deforestation, and achieved F1 score improvements of between 0.066 and 0.234 over existing methods. This was achieved primarily through a higher recall over other models, while using only 473 K parameters. SHAZAM demonstrates superior mapping capabilities through a higher spatial resolution and the suppression of background features for both immediate and gradual hazards. SHAZAM offers an effective and generalisable solution for hazard detection and mapping across different geographical regions and a diverse range of hazards, and highlights the potential of further self-supervised method contributions for hazard detection and mapping.https://ieeexplore.ieee.org/document/11033198/Anomaly detectionchange detectiondeep learninghazard detectionsatellite image time series (SITS)self-supervised learning
spellingShingle Samuel Garske
Konrad Heidler
Bradley Evans
Kee Choon Wong
Xiao Xiang Zhu
SHAZAM: Self-Supervised Change Monitoring for Hazard Detection and Mapping
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Anomaly detection
change detection
deep learning
hazard detection
satellite image time series (SITS)
self-supervised learning
title SHAZAM: Self-Supervised Change Monitoring for Hazard Detection and Mapping
title_full SHAZAM: Self-Supervised Change Monitoring for Hazard Detection and Mapping
title_fullStr SHAZAM: Self-Supervised Change Monitoring for Hazard Detection and Mapping
title_full_unstemmed SHAZAM: Self-Supervised Change Monitoring for Hazard Detection and Mapping
title_short SHAZAM: Self-Supervised Change Monitoring for Hazard Detection and Mapping
title_sort shazam self supervised change monitoring for hazard detection and mapping
topic Anomaly detection
change detection
deep learning
hazard detection
satellite image time series (SITS)
self-supervised learning
url https://ieeexplore.ieee.org/document/11033198/
work_keys_str_mv AT samuelgarske shazamselfsupervisedchangemonitoringforhazarddetectionandmapping
AT konradheidler shazamselfsupervisedchangemonitoringforhazarddetectionandmapping
AT bradleyevans shazamselfsupervisedchangemonitoringforhazarddetectionandmapping
AT keechoonwong shazamselfsupervisedchangemonitoringforhazarddetectionandmapping
AT xiaoxiangzhu shazamselfsupervisedchangemonitoringforhazarddetectionandmapping