نوع مقاله : علمی - پژوهشی
نویسندگان
1 استاد گروه سنجش از دور و GIS، دانشکدۀ علوم زمین، دانشگاه شهید بهشتی
2 استاد پژوهشکدۀ مدیریت خطرپذیری، پژوهشگاه بینالمللی زلزلهشناسی و مهندسی زلزله
3 مربی گروه سنجش از دور و GIS، دانشکدۀ علوم زمین، دانشگاه شهید بهشتی
4 کارشناس ارشد سنجش از دور و GIS، گروه سنجش از دور و GIS، دانشکدۀ علوم زمین، دانشگاه شهید بهشتی
چکیده
کلیدواژهها
عنوان مقاله [English]
نویسندگان [English]
Earthquake is one of the most destructive natural disasters which frequently occurs with different intensities. Earthquakes cause severe damage to buildings, main roads and most importantly, loss of life. Detection of damaged buildings caused by such an event at the right time is a critical issue for crisis management and disaster relief. The aim of this study is to detect earthquake damaged buildings using very high resolution (VHR) satellite imagery. To achieve this result, the satellite images with very high resolution before and after the earthquake in Port-au-Prince in Haiti as well as the observed destruction map in 2010 were used. In this study, the optimum features extracted from the image were selected using correlation analysis. The buildings destroyed were classified using fuzzy inference system and the values of selected textures. Finally, the damage map obtained from the proposed algorithm was compared to the map of the area. The kappa criterion estimated from the results of the proposed method is 82% while the index- Jaccard parameter is 89.69%.
کلیدواژهها [English]