New frontier of multi-hazard damage detection based on satellite SAR images taken after natural disasters

2020 Fiscal Year Final Research Report

• Project/Area Number: 17H02066
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Natural disaster / Disaster prevention science
• Research Institution: National Research Institute for Earth Science and Disaster Prevention (2019-2020); Chiba University (2017-2018)
• Principal Investigator: Yamazaki Fumio 国立研究開発法人防災科学技術研究所, 災害過程研究部門, 主幹研究員 (50220322)
• Co-Investigator(Kenkyū-buntansha): 劉 ウェン 千葉大学, 大学院工学研究院, 助教 (60733128)
• Project Period (FY): 2017-04-01 – 2021-03-31
• Keywords: リモートセンシング / 合成開口レーダ / 地震被害 / 風水害 / 地盤災害 / 被害把握 / 構造物被害 / 高分解能SAR衛星

Outline of Final Research Achievements

Synthetic Aperture Radar (SAR) has a scheme to emit microwaves to the earth surface and then observes the reflected echoes. SAR sensors onboard satellites or aircraft can be used day-time and night-time and in all weather conditions. In this research, damage detection of man-made structures and extraction of flooded/landslide areas due to natural disasters were carried out using post-event high-resolution SAR images and pre-event optical images and GIS data. Building damage due to the 2016 Kumamoto earthquake, flooding areas due to the 2018 Western Japan torrential rain and the 2019 Typhoon Hagibis, landslides due to the 2018 Hokkaido-Eastern-Iburi earthquake were extracted successfully using satellite SAR data. But the necessity of pre-event SAR data or land-cover and GIS data is suggested to avoid false extraction.

Free Research Field

防災リモートセンシング

Academic Significance and Societal Importance of the Research Achievements

衛星や航空機に搭載された高分解能の合成開口レーダ（SAR)画像を用いて，自然災害による構造物被害，浸水範囲，土砂災害などを早期に検出する手法を試みた．災害前後のSAR画像の変化抽出による方法に加えて，同じ条件で撮影した事前SAR画像が存在しない場合は，事前の土地被覆分類や地理空間情報を組み合わせによる災害検知の可能性を明らかにした．成果は，我が国や世界において，広域災害発生時の情報収集に寄与することが期待される．