Fast and progressive reconstruction method of point cloud based on star hypothesis

Abstract

The invention discloses a fast and progressive reconstruction method of a point cloud based on a star hypothesis, which can not only robustly and quickly reconstruct a three-dimensional model which closely surrounds the point cloud, is watertight, has no self-intersection and has a smooth surface, but also realize fast reconstruction for input changes such as local adding and deleting point cloud,a moving center point and the like. The method comprises the following steps of: dynamically collecting a point cloud of a reconstruction model of a reconstruction object; constructing an initial spherical model; projecting the point cloud onto the spherical surface of the initial spherical model and obtaining corresponding projection points, and calculating the mesh vertices closest to the projection points; calculating the target position of each mesh vertice, transforming the initial spherical model according to the target positions to obtain a star-like model; constructing a tar mesh constraint which is smooth and tightly surrounds the point cloud; recalculating the coordinates of the mesh vertices and transforming the star-like model to a star model; continuously inputting a new point cloud, updating the spherical center of the star model and continuously reconstructing the star model, so as to obtain the reconstruction model of the reconstruction object.

Description

technical field [0001] The present invention relates to the technical field of geometric figure processing, and specifically relates to but not limited to a fast and progressive point cloud reconstruction method based on a star assumption, so as to realize the fast and progressive configuration of point cloud data. Background technique [0002] The progressive point cloud reconstruction method is a method for quickly generating a 3D model from dynamic point cloud data. It is a point cloud-based 3D reconstruction research application for virtual environment simulation, mobile robot navigation, and medical 3D image reconstruction. At present, the commonly used progressive point cloud reconstruction methods are mainly divided into three types. The specific contents and limitations of these three methods are as follows: One is the implicit reconstruction method, which needs to solve the implicit function and extract the surface grid, but the calculation efficiency is low, so It ...

AI Technical Summary

Problems solved by technology

At present, the commonly used progressive point cloud reconstruction methods are mainly divided into three types. The specific contents and limitations of these three methods are as follows: One is the implicit reconstruction method, which needs to solve the implicit function and extract the surface grid, but the calculation efficiency is low, so It is not suitable for progressive reconstruction, and this method usually requires the input of the normal information of the point cloud, and it is difficult to estimate the normal direction during the progressive reconstruction process; the second is the explicit reconstruction method, such as the reconstruction method based on Alpha Shape, it is difficult to maintain the model The surface is smooth and prone to hole structures, and it will take time to improve the parameters if adaptive adjustment is used; the third is the method based on deformation, and uses point cloud data to drive the deformation of the initial mesh model, so that the mesh model closely surrounds the points cloud, but this method cannot avoid the self-intersection phenomenon of grid points and the corresponding numerical problems during operation

In addition, most of the existing methods can only reconstruct the point cloud collected on the surface of the reconstructed object, but cannot be directly applied to the point cloud collected inside the reconstruction

Images