Detecting collapsed buildings from image data for estimation of disaster debris

3rd International Conference on Energy Materials and Environment Engineering

能源学;生态环境科学

Yoshinari, N.^1,2,3 ; Yamamoto, M.^1,2,3 ; Tonegawa, R.^1,2,3 ; Oyanagi, Y.^1,2,3 ; Ohuchi, A.^1,2,3,4

Hokkaido University, Sapporo, Japan^1

Sumitomo Electric Industries Ltd, Osaka, Japan^2

Public Interest Incorporated Foundation for International Peace, Tokyo, Japan^3

Sendai Environmental Development Inc., Sendai, Japan^4

关键词: Aerial Photographs;    Affected area;    Collapsed buildings;    Image data;    Large amounts;    Temporary storage;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/61/1/012163/pdf
DOI  :  10.1088/1755-1315/61/1/012163

学科分类：环境科学（综合）

来源: IOP

【 摘 要 】

In the event of a massive earthquake, it is difficult to assess the situation in affected areas because access to these areas is usually blocked off. Tsunamis, in particular, create a large amount of debris and block many access paths. It is necessary to estimate the amount of disaster debris as early as possible in order to begin restoration in affected areas. This data helps to determine the required amount of temporary storage and arrange proper transportation. We propose a system that detects collapsed buildings after a disaster by using three aerial photographs. We verify the accuracy of the proposed method and discuss how detecting collapsed buildings aids in estimating total disaster debris.