Detection of Bridge Damages due to Tsunami Using Multi-temporal High-resolution SAR Images

Abstract

This study tried to detect collapsed bridges due to tsunami by analyzing changes of backscattering intensity in Synthetic Aperture Radar (SAR) images. TerraSAR-X images covering the Sendai and Ishinomaki planes acquired before and after the 2011 Tohoku-Oki earthquake were employed together with the GIS data of bridges. The bridge region was created according the GIS data with a 10-m buffer. Then the difference and correlation coefficient of the sigma naught values within the bridge region were obtained. The proper threshold value of the correlation coefficient was investigated to extract collapsed bridges. In order to verify the effectiveness of the method, the extracted results were compared with the reference data from damage investigation reports and aerial photographs. The threshold vale of the correlation coefficient, 0.47, gave the highest value for the summation of the producer accuracies for collapsed and survived bridges, with the overall accuracy 0.776 and the kappa coefficient 0.430.