2017 Fiscal Year Final Research Report
Development of automated terrain classification for global disaster prevention using medium to high resolution DEM
Project/Area Number |
15K01176
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Multi-year Fund |
Section | 一般 |
Research Field |
Geography
|
Research Institution | Geospatial Information Authority of Japan (Geography and Crustal Dynamics Research Center) |
Principal Investigator |
Iwahashi Junko 国土地理院(地理地殻活動研究センター), その他部局等, 主任研究官 (90391698)
|
Co-Investigator(Kenkyū-buntansha) |
神谷 泉 国土地理院(地理地殻活動研究センター), その他部局等, 技官 (80370815)
|
Co-Investigator(Renkei-kenkyūsha) |
MATSUOKA Masashi 東京工業大学, 大学院総合理工学研究科, 教授 (80242311)
|
Research Collaborator |
YAMAZAKI Dai 東京大学, 生産技術研究所, 准教授 (70736040)
Yong Alan US Geological Survey, Research Geophysicist
Herrick Julie US Geological Survey, Geologist and Map Editor
Mital Utkarsh California Institute of Technology, Postdoctoral Scholar
|
Project Period (FY) |
2015-04-01 – 2018-03-31
|
Keywords | 地形分類 / DEM / 機械学習 / クラスタリング / 階層クラスタリング / 地形 / 地質 / Vs30 |
Outline of Final Research Achievements |
A method to create polygon-based terrain classification data was developed and global terrain classification data by region segmentation, machine learning and hierarchical clustering were completed. Terrain groups with different topographical and geological features were detected by comparing with existing geological and geomorphological classification maps in Japan. The classification is improved from previous studies that used 1-km DEMs, especially for the representation of terrace shapes and landform elements smaller than 1 km. The results were generally suitable for distinguishing bedrock mountains, hills, large highland slopes, intermediate landforms (plateaus, terraces, large lowland slopes), and plains. Furthermore, Vs 30 (average shear wave velocity for the top 30 m) data were compared with the terrain classification data and showed the possibility of utilization for seismic hazard.
|
Free Research Field |
数値地形解析
|