Simplification Of 3D Laser Point Cloud Data Based On Curvature Estimation

With the development of science and technology,three-dimensional laser scanning technology appeared in the 1990 s,and this technology has been the focus of scholars at home and abroad.The 3D laser scanning technology can quickly,conveniently and accurately acquire the 3D information of the surface of the object,and the obtained point cloud data can be processed to construct a 3D model reflecting the real information of the object.Due to its non-contact,high efficiency,real-time and high precision,3D laser scanning technology has been widely used in various fields.With the increasing requirements for point cloud accuracy in practical applications,the large amount of point cloud data and high density have become the bottleneck of the development of 3D laser scanning technology.How to quickly and accurately simplify point cloud data has become the focus of 3D laser scanning technology research.In order to simplify the point cloud data,this paper introduces the 3D laser scanning technology and its data processing method in depth.Combined with the existing methods,the sub-regional point cloud simplification method is proposed to simplify the point cloud data.The following is the main work of this article:(1)This paper summarizes the research status of 3D laser scanning technology and point cloud simplification technology at home and abroad,introduces the principle and application of 3D laser scanning technology,and analyzes the error of 3D laser scanning system,and also introduces point cloud data.The acquisition and preprocessing process is focused on the principle and common methods of point cloud registration and point cloud denoising in point cloud data preprocessing.(2)This paper focuses on several classic simplified methods in point cloud data simplification,including random sampling method,uniform grid method based on bounding box,curvature sampling method,minimum distance simplification method and clustering method,and summarizes its advantages and disadvantages.(3)For the case where the point cloud distribution in the large scene is uneven and the different parts simplify the quality requirements,this paper proposes a sub-regional simplification method.The basic idea is to estimate the curvature of each point in the point cloud data.The cloud data is divided into flat regions and complex regions.The random region is simplified by the random sampling method with reserved boundaries.The complex region is simplified by the hierarchical curvature method.Finally,the point cloud data is merged and simplified.(4)Based on the MATLAB platform,the sub-regional simplified program was written.Combined with the point cloud data scanned by the field,the point cloud data was simplified.Finally,in the point cloud simplified evaluation,from the point cloud visual comparison,modeling visual comparison and modeling surface area change,compared with other point cloud simplification methods,the evaluation of the simplified quality of the point cloud is completed.Sub-regional point cloud simplification effect is better,simplifying higher quality.