Development of assistive technology to automate building damage assessment of tsunami flooded area

• Project/Area Number: 26750133
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Natural disaster / Disaster prevention science
• Research Institution: Tohoku University
• Principal Investigator: Shibayama Akihiro 東北大学, 災害科学国際研究所, 准教授 (80455451)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Project Status: Completed (Fiscal Year 2016)
• Budget Amount: ¥4,030,000 (Direct Cost: ¥3,100,000、Indirect Cost: ¥930,000); Fiscal Year 2016: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000); Fiscal Year 2015: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000); Fiscal Year 2014: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
• Keywords: 津波被害 / 建物被災度判定 / 全周囲カメラ / 機械学習 / 津波 / 建物被害 / 画像認識

Outline of Final Research Achievements

This research clarifies the technical method used outside the disaster area to confirm the building damage levels based on photographs of the tsunami disaster area taken using an in-vehicle all-around camera.To improve damage assessments, through making the evaluation of building damage level more automatic using the in-vehicle all-around camera, a method was clarified to present rough classifications of levels of disaster damage.Furthermore, in order to improve the accuracy of automatic evaluation, a mechanism was constructed that uses the visually observed evaluation results as training data of machine learning (Deep Learning), and a system was created to easily access this information on the web.