Building detection with LiDAR point clouds based on regional multi-return density analyzing

A new strategy for the classification of raw LiDAR points and building detection in urban areas is proposed based on the regional multi-return density analysis and which is constructed on the comprehensive utilization of echo features of different object types and terrain information. The main procedures of the classifying of the off-terrain points is beginning at the construction of Triangulated Irregular Network (TIN), then the regions of each object are captured by the contours clustering based on the topological relations of each contours traced from the TIN. Finally, the type of the object is recognized by the statistical analysis of the regional multi-return density for the significant difference on the building region and vegetation regions. This method makes good use of the difference on echo features of different objects such as buildings and trees but obeying the appearance of the multi-returns happened on the edges of the building. At the same time, the adaptive region determination of the objects is accomplished following the contours clustering. So the proposed method can dramatically increase the classification accuracy and overcome the weakness of the traditional methods which is more useful for the continued researches and applications such as building reconstruction and the parameters estimation of the tress. The experiment proves that the new algorithm can get an effective classification.