Study On Data Post-processing Method Of Airborne LiDAR System

Airborne LiDAR measurement technology is a new technology of surveying and mapping field. Airborne LiDAR measurement is not limited by the sunshine and the weather. It can directly obtain instantly 3D accurate information of the earth surface andobjects. So it has been widely used in disaster monitoring, protection of cultural relics, power line reconnaissance and designing, exploration of resources, forest survey, topographic surveying and mapping and some other fields. With the development of the airborne LiDAR measurement technology and application scope expands increasingly, most of the hardware technology and system integration problems have been resolved. But, point cloud data post-processing is current research hotspot and keynote in the field of surveying and mapping, that main contents boresight misalignment error calibration,point cloud automatic filtering, extracted topographic data from LIDAR data, and so on.For airborne LiDAR system and its data processing method has carried on the related research in this paper. First, analysed of the expounds composition of airborne LiDAR system and geometric positioning principle, researched the basic process of airborne laser radar data processing and the error sources of airborne LiDAR. This paper studies the theory of boresight misalignment calibration method.What’more, through the calibration experiment and its accuracy analysis verified the simply select characteristic features, and does not need to deploy ground control points of the correctness of the boresight misalignment error calibration method. In view of the current filter algorithm requires the characteristics of more parameters,high-threshold setting and poor adaptability, a filtering method are proposed in this paper based on the combination open operator of mathematical morphology with skewness balancing method. This method overcomes the faults of skewness balancing method need terrain unique degrees flat, makes full use of the degree of skewness balancing method does not need the threshold and take advantages the mathematical morphology open operator’s threshold setting is easy and high efficiency. Then, verified the feasibility of this method by experiment. For the ground point obtained through filter classification(housing, vegetation, etc), put forward a improved method of roof point extraction by the fuzzy cluster analysis(FCM). This method mainly overcomes that shortage of fuzzy cluster analysis of classification results related to the initial position and easy to have a local optimum.In addition, combining with simulated annealing algorithm and genetic algorithm, make it mutual complement each other, overcome the premature phenomenon of genetic algorithm effectively. So that the fuzzy cluster analysis is more efficient, converge to the global optimal solution more quickly, extracted roof point successfully.