Building Detection with Spatial Voting and Morphology Based Segmentation

Automated object detection in remotely sensed data has gained wide application areas due to increased sensor resolution. In this study, we propose a novel building detection method using high resolution DSM data and true orthophoto image. In the proposed method, DSM feature points and NDVI are obtained. Then, they are used for spatial voting to generate a building probability map. Local maxima of this map are used as seed points for segmentation. For this purpose, a morphology based segmentation method is proposed. This way, buildings are detected from DSM data. We tested our method on ISPRS semantic labeling dataset and obtained promising results.