Automated Detection of Damaged Buildings due to Earthquakes using Aerial HDTV and Photographs

We studied the possible use of aerial high definition television (HDTV) images taken after the 1995 Kobe earthquake, to establish a methodology for an automated detection of building damage. The relationship between the degree of building damage and the color and edge indices from the aerial images were examined by image processing techniques. The characteristics of building damage were defined on the basis of hue, saturation, brightness and edge intensity. Using a threshold value of these parameters, the study areas were classified into damaged and undamaged pixels. A filter analysis was further conducted to these pixels and damaged buildings were identified. The estimated damage distribution by the proposed methodology agrees well with the field survey data and the visual inspection of the aerial HDTV and photographs.