"Natural hazard is a natural phenomenon, which causes huge loss of lives, heavily infrastucture damages and great financial losses every year all around the globe. Among the natural hazards, earthquake is the most crucial one that effect the people and nature adversely. Four of every five deaths caused by earthquakes occurred in developing countries in the 20th century. More than a million earthquakes occur which is equal to two earthquakes per minute around the world according to the statistical data.Turkey is home to major earthquakes due to its tectonic structure. On October 23th 2011, a great earthquake occured in the eastern part of Turkey in Van Province. After the earthquake, the official death toll stands at 604, with some 4152 people injured, nearly 4,000 homes either damaged or collapsed. Earthquake-induced building damage detection is a very important step after earthquakes since earthquake-induced building damage is one of the most critical threats to cities and countries in terms of the area of damage, rate of collapsed buildings, the damage grade near the epicenters and also building damage types for all constructions.The motivation of this paper is to detect and extract the collapsed residential buildings and debris areas with object based image analysis and to compare the classification results based on orthophoto and Quickbird 2 data. In this research, two different land surfaces were selected as homogenous and heterogeneous case study areas.The first step of application was multi-resolution segmentation which was applied and optimum parameters were selected to obtain the objects in each area after testing different color/shape and compactness/smoothness values. In the second phase, nearest neighborhood classification procedure were applied and their classification performances were compared. Seven land cover types such as; vegetation, road, agricultural land, mixed area, buildings, collapsed buildings and as well as debris were successfully extracted. Error matrices, and the Kappa index of agreement (KIA) was selected as measures for accuracy evaluation. In the accuracy assessment analysis, first, control objects were selected for the two types of classification results applied in 4 different regions. According to the error matrix based on the control objects, the overall accuracy for the NN classification was found as 93% for the homogeneous region 1, with KIA as 0.90, and for homogenous region, the overall accuracy for the NN classification was found as 80% with KIA as 0.77. For the heterogenous regions as 1 and 2, the overall accuracy for the NN calssification was found 94% with KIA 0.93 and 66% with KIA 0.61 respectively. Promising results were obtained with both orthophoto and QB2 based classification.The Object-Based Image Analysis (OBIA) was performed using e-Cognition software."