About identifying man-made objects from mobile LiDAR data

A new automated approach for the detection and classification of man-made objects in urban corridors from point clouds acquired by vehicle-borne Mobile Laser Scanning (MLS) has recently been suggested by Fan and Yao. The approach is designed based on a-priori knowledge in urban areas:
(i) man-made objects feature geometric regularity like vertical planar structures (e.g. building facades), while vegetation reveals a huge diversity in shape and point distribution;
(ii) different types of urban man-made objects can be characterized by the point extension and height above the ground.
Therefore, MLS-based point clouds are first divided into three layers with respect to the vertical height dimension. In each layer, seed points of man-made objects are indicated by a line-filter in the footprints of off-ground objects which is generated by binarizing the spatial accumulation map of pointclouds.
These seed points are further classified by examining in which layers the seed points of objects are found. Finally, points belonging to respective objects can be retrieved based on classified seed points. The experiments show that various man-made objects on both street sides can be well detected with a detection rate of up to 83%. For the classification of detected urban objects, an overall accuracy of 92.37% can be achieved.