P2.30s Building damage estimation after natural disaster using multi satellite source data based on machine learning

Advanced Satellite Systems, Sensors & Intelligence

Chang Liu, The University of New South Wales

In order to have a fast post-event damaged building classification method, this research develops a deep learning method for categorizing damages into four levels, no damage/minor damage /major damage/collapsed. The three key objectives are to develop an accurate detection method for building detection using both 2D and 3D data; to specify more detailed damage categories based on building damage detection than only collapsed or not; and to develop a post event database based on building damage detection.