Fast processing of discrete data based on dynamical regular grid nets

With the development of LIDAR, InSAR and other new technologies used in the data acquisition, the three dimensional point cloud data has become a very important data source in the Geomatics applications. LIDAR provide a convenient way to acquire a massive three dimensional point data. A high efficiency algorithm for the management and searching of LIDAR data is an important foundation of the procedure of LIDAR processing such as filtering, display and threedimensional reconstruction. How to deal with a large amount of discrete data is not only one of the focuses and challenges in the area of LIDAR point cloud data processing, but also an important content in some other areas such as the DEM generation. While in some ways the storage, organization and management solutions of these massive points will affect the efficiency and accuracy in the following processing. In this paper a dynamical regular grid nets method will be introduced. This method will then be used in some airborne LIDAR filtering experiments of different areas. We will see the procedure and result of the point cloud data processing using this dynamic regular grid nets method. Accordingly, the effectiveness of this method will be proved.