Department of complexity science and engineering

Graduate School of Frontier Sciences

The University of Tokyo

Yoshikawa-Lab　Doctoral Course（first-year）

E-mail：fuda-fuca at g.ecc.u-tokyo.ac.jp

Room：新領域基盤棟4E5号室（柏キャンパス）

Research Themes：Machine learning based satellite image analysis and land change information extraction, Disaster monitoring using small satellite and nanosatellite data

Master’s Thesis Title：Toward Faster and Accurate Post-disaster Damage Assessment: Development of End-to-End Building Damage Detection Framework with Super-resolution Architecture

Research Summary：

1、Application of super-resolution techniques in remote sensing image analysis

2、A framework for fast building damage detection by fusing pre- and post-disaster images of different resolutions (based on end-to-end training)

Figure: Superiority of building damage detection accuracy and super-resolution quality demonstrated by the proposed approach (EEBSR-B) on the tsunami dataset

3、Plug-and-play module improves the accuracy of building damage detection by fusing features from pre- and post-disaster remote sensing images

Research achievements：

[1] Fu, Xuanchao, et al. “Toward Faster and Accurate Post-Disaster Damage Assessment: Development of End-to-End Building Damage Detection Framework with Super-Resolution Architecture.” IGARSS 2022-2022 IEEE International Geoscience and Remote Sensing Symposium. IEEE, 2022.