Geometric Feature Extraction Of Sea Dike Disaster-bearing Body Based On Mobile LiDAR

Various and frequent marine disasters have caused many unfavorable effects on coastal economic development and marine ecological environment.The sea dike disasterbearing body is one of the important marine disaster-bearing bodies.The investigation on the status of the sea dike disaster-bearing body is an important basic work to strengthen the capacity of building the marine disaster prevention and mitigation system,which can guarantee the people's lives and property,promote the rapid economic development and improve the marine ecological environment,and provide the authorities concerned with abundant and accurate basic geographic information,and provide basic data support for decision-making deployment of disaster mitigation and prevention.At present,the survey of the sea dike disaster-bearing body mainly includes the general artificial measurement method based on total station and GPS-RTK,and the photographic measurement method based on high-resolution remote sensing image.The general artificial measurement method has high precision but low efficiency,and the photographic measurement method based on high-resolution remote sensing image has high efficiency but low precision.As a new high precision mobile measurement technology,mobile LiDAR measurement can collect three-dimensional information of the sea dike disaster-bearing body quickly and accurately,which is an effective method to investigate the disaster-bearing body.The point clouds obtained by mobile LiDAR has the characteristics of mass,high redundancy and diversity,etc.How to realize the rapid extraction of the geometrical features of the sea dike disaster-bearing body,which is the key research direction of the LiDAR point cloud processing in the sea dike.In view of the above analysis,this paper selected sea dike disaster-bearing bodies in Maoming and Zhanjiang of Guangdong Province as the research objects.Based on the mobile LiDAR point clouds and geometric feature of sea dike disaster-bearing body,the geometric feature model of sea dike disaster-bearing body was established.Transform the point clouds two or three-D space coordinates,combined with the RANSAC algorithm,an automatic extraction algorithm based on geometric feature model of sea dike disasterbearing body was proposed and implemented in mobile LiDAR point clouds.The automatic extraction of geometric feature and the survey elements of the sea dike disasterbearing body was realized,and the feasibility of the proposed algorithm was verified by experiments.The main points of this paper were as following:1.Based on a large number of field investigation data and the relevant specifications of sea dike construction in Guangdong Province.This paper sumed up the standard and nonstandard sea dike structure types in Maoming and Zhanjiang in Guangdong Province,analyzed the geometric features of the sea dike with different structure types,and established the geometric feature model of the sea dike disaster-bearing body.2.Based on the prior knowledge,the coarse classification of the mobile LiDAR point cloud was carried out,and the point clouds denoising was realized by the elevation distribution histogram.Aiming at the pre-processed LiDAR point clouds of the sea dike,the automatic extraction algorithm of the geometric feature of the sea dike disasterbearing body was proposed on the basis of the geometric feature model of the sea dike.3.The results of automatic extraction algorithm of the geometric feature of the sea dike disaster-bearing body and the artificial extraction results were compared to verify the feasibility of the algorithm.The following conclusions were obtained:(1)Automatic extraction algorithm the minimum RMS error of automatic extraction results was up to 0.094 for the standard structural type sea dike with a length of about 1.1km,and the minimum RMS error of the automatic extraction result was up to 0.000 for the non standard structure type sea dike of about 0.2km length.(2)In time efficiency,acquiring the survey elements of the same sea dike with a length of about 1.1km,the automatic extraction algorithm saves about 80.95% of time and greatly improves the working efficiency compared with the manual extraction method.