Research On 3D Point Cloud Feature Extraction And Optimization Method Based On Morse Comple

Topological feature extraction from 3D point clouds is a key and fundamental step in 3D model construction,widely used in building modeling,3D landscape display,virtual reality,and other fields.However,the process of extracting model topological features based on traditional algorithms still faces problems such as excessive redundant information and lack of topological feature representation information.Therefore,it is of great significance to study how to reduce the amount of raw data and improve the integrity,topology consistency,and accuracy of point cloud topology feature extraction in 3D models through existing computer software and hardware resources.In response to the above issues,this study systematically compared and analyzed the advantages of using relative normal vectors and second-order microcomponent curvature as functional indicators for feature recognition through experiments.This study discusses how to optimize feature extraction methods based on calculating the trend of surface changes around vertices.Curvature is used to characterize the key point selection index of the Morse function,and a new method for calculating the importance metric value is constructed to determine the importance of key lines in complex networks.Optimization methods can extract more accurate and complete complex critical networks,effectively reducing the amount of data in point cloud models.Finally,the relevant algorithms and models for topology feature extraction and simplification are integrated into an experimental system.The main research work is as follows:(1)This study summarizes the problems faced by current 3D point cloud feature extraction,and discusses the advantages and disadvantages of mainstream 3D point cloud feature extraction methods at home and abroad.(2)In the process of feature extraction,there are limitations such as complex vertex recognition,excessive computational complexity,and rough fitting of vertex surface morphology.This study improves the key point selection index by selecting the second-order derivative curvature of vertices as the Morse function index.The curvature calculation process is relatively simple and requires a small amount of calculation,which can better fit the change trend of bumps and depressions on the surface of the point cloud data model.The experimental results show that using curvature to characterize the Morse function selection index can better identify and select key edge feature points of the point cloud model,laying a good foundation for the construction of the complex edge network of the model.(3)A method for calculating the standard value of importance measurement has been constructed to simplify the key lines of complex networks.Feature keylines are mainly composed of key vertices and line segments between them.The critical duration method only considers the Morse function values of the vertices on the feature lines of the complex network,without paying attention to another important component of the feature key line-line segment length.This method extracts too many feature key line branches in the process of simplifying the complex marginal network,and the identified model edge contour lines exhibit fracture phenomenon.Therefore,This study constructs a new feature importance measurement method.The new method comprehensively considers the index value of Morse function and the length of line segments to simplify the complex marginal network.The experimental results demonstrate that the new calculation method has a more significant effect in removing feature key line branches and extracting important model feature lines.(4)In order to explore the scope of 3D models that optimization methods can adapt to,this article uses two big data scale models with different component modules for analysis.Among them,the component modules and construction method of the Xuanyuan Gate model are relatively simple,while the component module types of the Parliament Building model are relatively complex.The results indicate that the optimization method has high applicability for constructing simple 3D point cloud models with smooth surfaces.It can effectively remove surface redundant feature key lines and extract model framework structural lines with good results.For models with complex and uneven surface components,this method extracts more complete edge edges and corners,but still leaves some redundant key additional lines.Future research will combine the advantages of other methods to explore the next step around this issue,so as to improve the extraction performance of complex models.(5)In order to facilitate the automated operation of 3D point cloud feature extraction and simplification in practical production,an experimental system for 3D point cloud feature extraction was designed and developed using object-oriented programming ideas.This study employs the MVC architecture design pattern to integrate the relevant algorithms and models for point cloud topology feature extraction and simplification into a system.The system is clearly divided into modules such as business logic,data processing,and interface display,which has functions such as file import and save,feature extraction and topology simplification,data statistics,and result display.