首页|Investigators at Delft University of Technology Detail Findings in Machine Learn ing (Integrating Post-event Very High Resolution Sar Imagery and Machine Learnin g for Building-level Earthquake Damage Assessment)
Investigators at Delft University of Technology Detail Findings in Machine Learn ing (Integrating Post-event Very High Resolution Sar Imagery and Machine Learnin g for Building-level Earthquake Damage Assessment)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Investigators publish new report on Ma chine Learning. According to news reporting from Delft, Netherlands, by NewsRx j ournalists, research stated, "Earthquakes have devastating effects on densely ur banised regions, requiring rapid and extensive damage assessment to guide resour ce allocation and recovery efforts. Traditional damage assessment is time-consum ing, resource-intensive, and faces challenges in covering vast affected areas, o ften limiting timely decision-making." Financial support for this research came from Netherlands Organization for Scien tific Research (NWO). The news correspondents obtained a quote from the research from the Delft Univer sity of Technology, "Space-borne synthetic aperture radars (SAR) have gained att ention for their all-weather and day-night imaging capabilities. These advantage s, coupled with wide coverage, short revisits and very high resolution (VHR), ha ve created opportunities for using SAR data in disaster response. However, most SAR studies for post-earthquake damage assessment rely on change detection metho ds using pre-event SAR images,which are often unavailable in operational scenar ios. Limited studies using solely post-event SAR data primarily concentrate on c ity-block-level damage assessment, thus not fully exploiting the VHR SAR potenti al. This paper presents a novel method integrating solely post-event VHR SAR ima gery and machine learning (ML) for regional-scale post-earthquake damage assessm ent at the individual buildinglevel. We first used supervised learning on case- specific datasets, and then introduced a combined learning approach, incorporati ng inventories from multiple case studies to assess generalisation. Finally, the ML model was tested on unseen study areas, to evaluate its flexibility in unfam iliar contexts. The method was implemented using datasets collected during the E arthquake Engineering Field Investigation Team (EEFIT) reconnaissance missions f ollowing the 2021 Nippes earthquake and the 2023 Kahramanmaras earthquake sequen ce. The results demonstrate the method's ability to classify standing and collap sed buildings, achieving up to 72% overall accuracy on unseen regi ons."
DelftNetherlandsEuropeCyborgsEme rging TechnologiesMachine LearningDelft University of Technology
NETL
NSTL
