| In recent years,with the development and widespread use of high-precision and highresolution Li DARs,3-D laser point cloud has increasingly gained popularity in many social fields such as medical care,agriculture,unmanned driving,geological prospecting,and cultural relics protection.More and more widely used.In addition to the spatial coordinate information of the object,the 3-D laser point cloud can also contain the reflection intensity information of the object,topological structure information,etc.Therefore,compared with the image,the 3-D laser point cloud can more intuitively express the characteristics of the object in details,which can satisfy The needs of a variety of different application scenarios.How to correctly represent the physical model of space objects based on three-dimensional point cloud data and realize the point-to-surface representation of space objects is a current research hot issue.Since the 1980 s,a large number of scholars have contributed new concepts and new algorithms to promote the continuous and in-depth development of surface reconstruction of 3-D point cloud.Until today,the surface reconstruction algorithms of 3-D point cloud are mainly divided into two categories-explicit surface reconstruction and implicit surface reconstruction.Each type of algorithm has its own advantages and disadvantages when applied to different industrial scenes.In this project,two real teaching environments in the school’s experimental building are selected as the research objects.The radial basis function is used to continuously approximate the real surface of the object,and the moving cube algorithm is used to triangulate the reconstructed implicit surface.Finally,the mesh is used.The way of the model represents the physical model of the space object.The implicit surface reconstruction method is compared with the reconstruction results of the explicit surface reconstruction algorithm Alpha Shape.In the presence of noise point cloud interference,the method of this topic is better than the Alpha Shape algorithm.How to calculate some characteristic parameters of space objects based on threedimensional point cloud data,such as volume parameters,is also a problem that needs to be faced in industrial production.For example,calculation problems such as the displacement of ships,the landfill volume of garbage dumps,and the storage volume of forest resources,are essentially the calculation of the volume parameters of these objects.At present,many scholars have also proposed different research methods for this problem: some directly calculate the volume of the object based on 3-D point cloud data;some first seek the convex hull model or mesh model of the object based on the 3-D point cloud data,and then based on the model Calculate the volume of the object.On the basis of surface reconstruction,the mesh model of the reconstructed object is further applied to the volume calculation of the object.The volume calculation method used in this subject directly manipulates each triangle facet,and calculates the volume of the model by projection and segmentation.The principle is simple and the realization is easy.Good 3-D point cloud data with complete data and no artifact point cloud interference is a prerequisite to ensure the correct results of the above processing.Therefore,this topic also introduces the preprocessing operation of 3-D point cloud,and discusses the influence of each parameter change in statistical filtering and voxel down-sampling on the volume calculation results.Finally,this subject proposes a smooth surface completion method based on a threedimensional point cloud mesh model for the problem of point cloud completion. |
PDF Full Text Request