Detection of collapsed building from unmanned aerial vehicle data with object based image classification

Buildings are most affected the objects by earthquake disaster. Detection of collapsed buildings after an earthquake isimportant both for determining the current situation and quick response. Unmanned aerial vehicles that have evolved inrecent years, can provide very high resolution images of the earth surface using camera systems attached to them.Information for the intended purpose can be obtained through the products produced from these images.In this study, collapsed buildings were detected in the area where high-resolution images were obtained whit unmanned aerialvehicle in 2015 and 2014. Building detection process was made based on a scenario events. In this context, 2015 images weretaken before the earthquake and 2014 images were taken after the earthquake. The images of both years were processedseparately to produce the digital elevation model and orthophoto image of the study area. building of the study area wereobtained by applying the object-based classification process to the generated data. 11 buildings which were available in thearea in 2015 and not available in the area in 2014, were detected successfully comparison of building classes of two years.