Water level estimation in flood area and stricken man-made objects detection using radar polarimetry

2015 Fiscal Year Final Research Report

• Project/Area Number: 25350495
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Natural disaster / Disaster prevention science
• Research Institution: Niigata University
• Principal Investigator: Sato Ryoichi 新潟大学, 人文社会・教育科学系, 教授 (00293184)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Keywords: 地震 / 津波・洪水 / 被災地観測 / レーダポーラリメトリ / リモートセンシング / 合成開口レーダ / 偏波散乱解析 / FDTD解析

Outline of Final Research Achievements

In this research, we propose useful techniques for accurately understanding the situation of stricken area, by making full use of quad-pol SAR data. Here, two useful algorithms are proposed as 1) Water level estimation algorithm by considering the distance between bridge and water surface in flooded area, and 2) Stricken man-made objects detection algorithm based on eigenvalue/eigenvector analysis. In both algorithms, polarimetric orientation angle compensation is carried out as a pre-processing to each main process. It was confirmed by the result of PolSAR image analysis for ALOS/PALSAR and ALOS-2/PALSAR-2 data that each algorithm is valid for the water level estimation and the man-made objects detection. Furthermore, it was verified for the quad-pol SAR data acquired by polarimetric scattering measurement for a scaled man-made object model with inclined sidewalls in anechoic chamber that the detection algorithm is still effective even for the deformed/inclined objects.

Free Research Field

波動情報工学