Yaser Hamednia - Ms student, International Institute of Earthquake Engineering and Seismology, Tehran, Iran
Babak Mansouri - Assistant Professor, Dept. of Emergency Management, International Institute of Earthquake Engineering and Seismology, Tehran. Iran
Kambod Amini-Hosseini - Associate Professor, Risk Management Research Center, International Institute of Earthquake Engineering and Seismology, Tehran. Iran

In disaster management operations right after disastrous earthquakes, damage maps areessential as they reveal the exact location of hard hit areas in urban settings. This study investigatesthe feasibility of an automated soft computing algorithm in generating damage maps. The main ideais to classify different areas associated to building footprints pronouncing three distinct damagelevels. A fuzzy inference methodology is exploited to determine the damage grade for each buildingroof by the means of evaluating the contribution of different identified damage classes. Forimplementation, satellite images of before and after the 2003 Bam, Iran earthquake, are used inaddition to some available ancillary data. The pre-processing step involved image coregistration andimage enhancement. Next, building footprint pixels were extracted from pre- and post-images usingthe ancillary building mask. Haralick’s second order textural features were computed for the buildingobjects and an optimum set of such features were selected using Genetic Algorithm (GA). Then,considering these optimal textural indices, different parts of individual roofs were classified intothree damage patterns as intact , partially-damaged and fully-damaged employing a SupportVector Machine (SVM) supervised classification algorithm. Fuzzy inference engine was used todetermine the damage grade of each building as to produce the damage map. The proposed algorithmwas evaluated by comparing the produced damage map with a reference damage map as ground truthwhere the results demonstrated the efficacy of the method showing an overall accuracy of 76% thatis suitable for rapid monitoring process.