Authors: Joshua Dimasaka, Christian Geiß, Emily So

Published on: April 02, 2024

Impact Score: 8.4

Arxiv code: Arxiv:2404.01748

Summary

• What is new: Introduction of OpenSendaiBench and the use of ResNet-50 models for mapping disaster risk from remote sensing data.
• Why this is important: Many countries cannot monitor climate and disaster risk effectively due to the high costs of large-scale surveys.
• What the research proposes: Using machine learning and time-series remote sensing data to map the distribution of exposure and physical vulnerability.
• Results: Successful mapping in Dhaka, Bangladesh demonstrates the potential for global disaster risk auditing over time.

Technical Details

Technological frameworks used: OpenSendaiBench

Models used: ResNet-50

Data used: Sentinel-1 SAR GRD and Sentinel-2 Harmonized MSI

Potential Impact

Disaster risk assessment firms, insurance companies, climate change mitigation businesses, NGOs, and governments in least developed countries.

Want to implement this idea in a business?

We have generated a startup concept here: RiskMapper.