Multi-sensor based image analysis for early and wide damage detection

2015 Fiscal Year Final Research Report

• Project/Area Number: 25282117
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Partial Multi-year Fund
• Section: 一般
• Research Field: Natural disaster / Disaster prevention science
• Research Institution: Tokyo Institute of Technology
• Principal Investigator: Matsuoka Masashi 東京工業大学, 総合理工学研究科, 准教授 (80242311)
• Co-Investigator(Kenkyū-buntansha): MIURA Hiroyuki 広島大学, 工学研究科, 准教授 (30418678) ; KOSHIMURA Shunichi 東北大学, 災害科学国際研究所, 教授 (50360847) ; SATOH Toshiaki 東京工業大学, 総合理工学研究科, 連携准教授 (50567146)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Keywords: 防災 / リモートセンシング / 自然災害 / 画像

Outline of Final Research Achievements

We developed an estimation model for calculating the water occupancy based on satellite SAR observation, comparing backscattering coefficients of the ALOS PALSAR image and the water fraction information derived from SMA method, which is applied to the Landsat-7 ETM+ reflectance image observing the 2011 Bangkok flood and the 2011 Tohoku earthquake tsunami.
We performed the three-dimensional building damage models, created by applying the SfM-MVS technology to images taken by the aerial photographs of the site of the Hiroshima heavy rain disaster in 2014, a three-dimensional model was generated and we clarified the relationship between the spatial resolution and the accuracy of the aerial photographs by comparison with the LiDAR data. Furthermore, a three-dimensional model was created from the images taken before and after the Great Hanshin-Awaji earthquake, and the building damage was extracted from the altitude differences in the buildings measured in these two different periods.

Free Research Field

防災リモートセンシング