Object extraction based on 3D-segmentation of LiDAR data by combining mean shift with normalized cuts: Two examples from urban areas

In this work, we have looked into the problem of urban analysis usingairborne LiDAR data based on the strategy of classification by segmentation.Segmentation is a key and hard step in the processing of 3D pointclouds, which is not perfectly solved in view of different applications.A new 3d segmentation method incorporating the advantages of nonparametricand spectral graph clustering is presented here to facilitate thetask of object extraction in urban areas. This integrated methodfeatures local detection of arbitrary modes and globally optimizedorganization of segments concurrently, thereby making it particularlyappropriate for partitioning raw airborne LiDAR data of urban areasinto segments approximating semantic entities. Two examples in urbanareas - flyover and vehicle are chosen as interest objects to beextracted by a classification-based step. The approach has been testedon LiDAR data of dense urban areas, and the results that are obtainedhave been compared with manual counts and showed us the efficiencyand reliability of the strategy.     «

In this work, we have looked into the problem of urban analysis usingairborne LiDAR data based on the strategy of classification by segmentation.Segmentation is a key and hard step in the processing of 3D pointclouds, which is not perfectly solved in view of different applications.A new 3d segmentation method incorporating the advantages of nonparametricand spectral graph clustering is presented here to facilitate thetask of object extraction in urban areas. This integrated methodfeatures local...     »