Determining the extent of building destruction after an earthquake using satellite imagery and fuzzy logic (Case study of Sarpol-e-Zahab region)

Earthquake is one of the most destructive natural disasters. Earthquakes occur in urban areas, destroying buildings and injuring people living in them. Urban center buildings are among the features that are exposed to many hazards during an earthquake. One of the first measures taken after an earthquake is relief. Locating damaged buildings can speed up relief efforts. One of the main sources in preparing the demolition of buildings is the science of remote sensing. Many researches have been done in the field of damage assessment using remote sensing images and the purpose of this research is to determine the extent of building destruction after an earthquake using plaidis data and fuzzy logic.

In this study, we used Plydez satellite images related to before and after the earthquake, in a period of less than one week after the earthquake. In this study, the first-order statistical tissue properties and the properties obtained from the matrix also were used. Then, using the genetic optimization algorithm, the optimal tissue properties were selected and using the isodita clustering and integration algorithm. Spectral information and optimal texture properties were determined for the roofs of the extracted buildings of three classes.

The condition of the buildings was done by a fuzzy inference system with Mamdani inference motor and fuzzy rules database designed using reference buildings and a total of 250 buildings were entered into the algorithm and classified into three classes of destruction.

The main output of this research was the demolition plan including three classes of healthy buildings to low degradation, high degradation and destruction. The use of reference buildings in estimating the accuracy of the proposed method showed an accuracy of 74% and acceptable method in determining the condition of buildings after the earthquake.