Object-based Earthquake-induced Damage Assessment Using Remote Sensing

The traditional application of aerial/satellite remote sensing data for earthquake-induced damage assessment use pixel-based change detection between pre- and post-event images. The limitation of this approach is in differentiating earthquake-induced damages from other natural and manmade changes in these images. In recent years, the object-based methods that are based on the concept that important semantic information is not represented in single pixel value alone but in meaningful image objects consisting of several pixels (i.e. context) has proved superior over pixel-based methods in several image processing applications. However, in order to effectively utilize the object based methods, we need a comprehensive understanding of potentially useful object characteristics and contexts of earthquake-induced damage. Through this research, it will be possible to analyze critical object characteristics of earthquake-induced damages that can be utilized for object-based damage assessment from optical and near-infrared aerial/satellite images from the 2010 Port-au-Prince, Haiti and 2011 Christchurch, New Zealand earthquakes.