نوع مقاله : مقاله علمی - پژوهشی
نویسندگان
1 دانشجوی دکتری سنجش از دور، دانشکده مهندسی نقشهبرداری، دانشگاه صنعتی خواجه نصیرالدین طوسی
2 استادیار، دانشکده مهندسی نقشهبرداری، دانشگاه صنعتی خواجه نصیرالدین طوسی
3 دانشیار، دانشکده مهندسی نقشهبرداری، دانشگاه صنعتی خواجه نصیرالدین طوسی
4 -کارشناسی ارشد مهندسی مکانیک، دانشکده مهندسی مکانیک، دانشگاه صنعتی خواجه نصیرالدین طوسی
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
Abstract: Fast relief and response is one of the responders’ tasks after the vigorous earthquakes in urban areas. Building damage maps can specify the relief and response priority. Remote sensing is one of the main sources in damage map production. Various research has been carried out on damage assessment using remote sensing data. In this research, the information about the classes has been extracted using pixel based classification from post-earthquake satellite image. By using the information about the classesand ANFIS model, a fuzzy system was designed that determines the rate of damage to the buildings using the percentage of the pixels engaged by building, shadow and debris in the building's polygon. The results of this study show that designed neuro-fuzzy system achieved the overall accuracy equal 0.92 in classifying buildings to the three damage level. Buildings were classified to undamaged, heavy damage and destroyed levels.
کلیدواژهها [English]