Development of Damage Detection Method for Wide and Gigantic Disaster Using Synthetic Aperture Radar (SAR)
| Project/Area Number |
17510155
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Natural disaster science
|
|---|
| Research Institution | National Research Institute for Earth Science and Disaster Prevention |
|---|
Principal Investigator |
MATSUOKA Masashi National Research Institute for Earth Science and Disaster Prevention, Earthquake Disaster Mitigation Research Center, Team leader (80242311)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
YAMAZAKI Fumio Chiba University, Graduate School of Engineering, Professor (50220322)
|
|---|
| Project Period (FY) |
2005 – 2006
|
| Project Status |
Completed (Fiscal Year 2006)
|
| Budget Amount *help |
¥3,600,000 (Direct Cost: ¥3,600,000)
Fiscal Year 2006: ¥1,100,000 (Direct Cost: ¥1,100,000)
Fiscal Year 2005: ¥2,500,000 (Direct Cost: ¥2,500,000)
|
|---|
| Keywords | Natural Disaster / Earthquake / Synthetic Aperture Radar / Satellite Image / Disaster Mitigation / Damage Detection / 合成開口レーダ / 人工衛星 |
|---|
| Research Abstract |
An automated damage detection method for use with remotely acquired synthetic aperture radar (SAR) images is proposed. Satellite-borne SAR can be used to detect the effects of large-scale disasters promptly following event, regardless of weather or sunlight conditions, and is thus potentially invaluable in recovery planning. The method proposed here involves the use of two pre-event SAR images and one post-event image. By calculating the difference between pre-and co-event correlation coefficients using the two pairs of images spanning the event, the effects of noise and surficial changes over time are minimized, allowing damage on smaller scales and in less densely developed areas to be resolved. The method is applied to evaluation of the extent and distribution of damage caused by the 2004 Niigata-ken Chuetsu earthquake in Japan (Mw 6.5) and the 2006 Central Java earthquake in Indonesia (Mw 6.3), which were observed by the Japanese ALOS and Canadian Radarsat satellites, respectively. The distributions of the difference value in built-up areas is in good agreement with damage reported in survey reports for both events. Large-scale landslides in are also resolved in some areas. The relationship between the difference value and the rates of severe damage to buildings is determined, and sub-district mapping of this relationships is demonstrated to correlate well with the damage statistics for the Central Java earthquake.
|
Report
(3 results)
Research Products
(23 results)
