English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Climate change results in an increased probability of extreme weather eventsthat put societies and businesses at risk on a global scale. Therefore, nearreal-time mapping of natural hazards is an emerging priority for the support ofnatural disaster relief, risk management, and informing governmental policydecisions. Recent methods to achieve near real-time mapping increasinglyleverage deep learning (DL). However, DL-based approaches are designed for onespecific task in a single geographic region based on specific frequency bandsof satellite data. Therefore, DL models used to map specific natural hazardsstruggle with their generalization to other types of natural hazards in unseenregions. In this work, we propose a methodology to significantly improve thegeneralizability of DL natural hazards mappers based on pre-training on asuitable pre-task. Without access to any data from the target domain, wedemonstrate this improved generalizability across four U-Net architectures forthe segmentation of unseen natural hazards. Importantly, our method isinvariant to geographic differences and differences in the type of frequencybands of satellite data. By leveraging characteristics of unlabeled images fromthe target domain that are publicly available, our approach is able to furtherimprove the generalization behavior without fine-tuning. Thereby, our approachsupports the development of foundation models for earth monitoring with theobjective of directly segmenting unseen natural hazards across novel geographicregions given different sources of satellite imagery.

Toward Foundation Models for Earth Monitoring: Generalizable Deep Learning Models for Natural Hazard Segmentation