Abstract
In response to the severe earthquakes in Türkiye in 2023, this study introduces the KATE-PD (KAhramanmaras Türkiye Earthquake Post Disaster) dataset, a valuable resource for the remote sensing and disaster response community. Unlike existing datasets, KATE-PD offers an unprecedented level of detail with high-quality annotated collapsed buildings across an area of 3,293 square kilometers, utilizing multi-source high-resolution satellite imagery. This dataset is particularly crucial for machine learning algorithms focused on damage detection from single post-earthquake images, enhancing the ability to assess damage quickly and accurately. The utilization of this dataset allows for significant improvements in the precision and speed of damage detection post-disaster, showcasing the critical role of well-curated datasets in advancing emergency management technologies. This version highlights the specific utility of the dataset for machine learning applications in detecting damage from single images, aligning with the needs of rapid response scenarios.
| Original language | English |
|---|---|
| Pages (from-to) | 1454-1457 |
| Number of pages | 4 |
| Journal | International Geoscience and Remote Sensing Symposium (IGARSS) |
| DOIs | |
| Publication status | Published - 2025 |
| Event | 2025 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2025 - Brisbane, Australia Duration: 3 Aug 2025 → 8 Aug 2025 |
Bibliographical note
Publisher Copyright:© 2025 IEEE.
Keywords
- deep learning
- Earthquake damage assessment
- machine learning
- remote sensing
- Türkiye Earthquake
Fingerprint
Dive into the research topics of 'A RAPID DAMAGE ASSESSMENT USING REMOTE SENSING: TURKIYE 2023 POST-EARTHQUAKE DATASET: KATE-PD'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver