Remote Sensing Extraction of Damaged Buildings in the Shigatse Earthquake, 2025: A Hybrid YOLO-E and SAM2 Approach

• Published in: Sensors (Basel, Switzerland) Vol. 25; no. 14; p. 4375
• Main Authors: Wu, Zhimin, Qu, Chenyao, Wang, Wei, Miao, Zelang, Feng, Huihui
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 12.07.2025; MDPI
• Subjects: Accuracy; Adaptation; Architecture; Buildings; China; Classification; damaged building extraction; Datasets; Disaster relief; earthquake damage assessment; Earthquakes; Forecasts and trends; multi-source remote sensing; Natural disaster damage; Neural networks; Post-disaster reconstruction; Remote sensing; Satellites; Segment Anything Model 2 (SAM2); Semantics; Topography; Turkey; YOLO-E; China; Turkey; earthquake damage assessment; multi-source remote sensing; Segment Anything Model 2 (SAM2); YOLO-E; damaged building extraction

In January 2025, a magnitude 6.8 earthquake struck Dingri County, Shigatse, Tibet, causing severe damage. Rapid and precise extraction of damaged buildings is essential for emergency relief and rebuilding efforts. This study proposes an approach integrating YOLO-E (Real-Time Seeing Anything) and the...