Providing Building Damage Map after Earthquake by Using Post-Event High-Resolution Satellite Imagery

Receiving rapid، accurate and comprehensive knowledge about the conditions of damaged buildings after earthquake strike and other natural hazards is the basis of many related activities such as rescue، relief and reconstruction. Fewer Restrictions by using high-resolution Images compared to terrestrial techniques turned it into the main source for damage assessment during these days. The present research aims at providing a damaged maps in a short time after the earthquake by using post-event high-resolution satellite Imagery. In this research، for all buildings، textural features for any candidate buildings، one by one، are extracted. Before extracting the features، optimum feature selection is done by genetic algorithm (GA). After selecting optimum textural features، buildings situation regarding to their destruction is evaluated using these features in two ''intact'' and ''Damaged'' classes by using the Maximum Likelihood and the Neural Network Algorithms. Finally، polygons of buildings classified in three different classes named ''Under 30%''، ''Between 30% and 70%'' and ''upper 70%'' based on the number of damaged pixels. The overall accuracy of classification evaluated about 81/04% by using an expert operator.