Accuracy Analysis And Evaluation Of Airborne LiDAR Data

Airborne Light Detection and Ranging (Airborne LiDAR) is regarded as one of the most revolutionary achievements for collecting remote sensing information as a new means of acquiring information in the recent decades. As the technology can quickly obtain the digital elevation model with high accuracyz and high spatial resolution. Thus, it has a unique advantage in the field of disaster monitoring, environmental monitoring, resource exploration, forest survey, topographic mapping and so on. Although the development of airborne LiDAR has nearly two decades, but the research of data processing is lagging behind. As a result, the most of existing research is still focused on how to design a reasonable and efficient filtering algorithm for the original point cloud.As the airborne laser radar system is composed of multiple parts, the acquire of point cloud data can be affected by many error sources which include GPS dynamic positioning error, INS attitude measuring error, laser scanning measuring system errors etc. The combined effects of various factors mainly reflected in the accuracy of point cloud data.The paper studys the problem of point cloud data precision after detailed analyzing the principle of airborne LiDAR and the characteristics of point cloud data. The paper analyses the accuracy of point cloud data from the several aspects of point cloud density, filtering method and interpolation method. It includes the following:â‘ It uses two methods to process the point cloud data and analyse the influence of DEM accuracy with different filtering methods.â‘¡It uses two types of data format to analyse the accuracy of point cloud data.One is discrete point cloud, another is DEM that is interpolated with discrete point cloud.â‘¢The research method of point-cloud data in precision is using the different number of ground inspections points to analyse the accuracy of point cloud data in the circumstances of different terrain types, different point cloud density and different interpolation methods. It gets the intentionally conclusions from analysing the accuracy of point cloud data with the different filtering methods, the different point cloud density, the different terrain conditions.