Based On Airborne LiDAR, Typical Coastal Feature Identification And Tidal Flat DEM Construction

The measurement of tidal flat terrain and the investigation of the types of coastal land use are of profound significance for understanding the dynamic changes of coastal topography and landforms,the rational development of tidal flat resources and the protection of coastal wetlands.Among them,the development and distribution of roads,buildings and vegetation can reflect the economic development of coastal zones,with special research value.As an emerging technology,the airborne LiDAR system can quickly acquire three-dimensional information on the ground surface.Compared with other conventional measurement methods,airborne LiDAR technology has unique advantages in tidal flat terrain survey and coastal land use type survey.However,the domestic LiDAR point cloud data processing method is still in its infancy.The existing data filtering and data interpolation algorithms have their own advantages and disadvantages.It is difficult for any one algorithm to adapt to all terrain.Coastal terrain features lack systematic methods and processes,so the research on this issue in this paper has important theoretical and practical significance.In this paper,the Yangkou Port area of Rudong County,Jiangsu Province is selected as the research area.The basic data is collected,including the point cloud raw data,and is developed for the gross error processing,data filtering,data interpolation and secondary classification in the point cloud data processing link A series of research and experiments,the main research work is as follows:(1)The composition and principle of the airborne LiDAR system and the data format of the point cloud are explained,and the technology is compared with the airborne In SAR technology and photogrammetry technology to analyze its advantages and disadvantages;(2)Through the auxiliary analysis of the original point cloud vignette and section map,the elevation statistical histogram is used to eliminate the rough points,which provides the basis for subsequent point cloud data processing;(3)By implementing three filtering algorithms: mathematical morphology,point cloud normal vector clustering,and irregular triangulation TIN,the point cloud data in the study area is filtered in order,and the filtering results are analyzed to compare the filtering accuracy of the three algorithms;(4)The inverse distance weighting,kriging and natural neighboring point interpolation methods are used to interpolate the filtered point cloud data to generate tidal flat DEM.The measured results of the research area are used to qualitatively and quantitatively evaluate the interpolation results of each method;(5)Based on Terra Scan software,systematically put forward the extraction method of typical features in the coastal zone(roads,vegetation and buildings),and further analyze the extraction results.The results of the study show that the number of coarse points in the mudflat area obtained by airborne LiDAR is small and can be eliminated welll.The irregular triangular mesh filtering algorithm is more adaptable to the terrain of the study area.After filtering and manual editing,the tidal flat DEMs generated by the three interpolation methods have higher accuracy,and the natural neighboring point interpolation method has higher interpolation efficiency and Elevation accuracy.Based on the remote sensing image and Terra Scan software,the results of road,vegetation and buildings are generally better.