Point cloud data quality automatic optimizing method and system

Abstract

The invention relates to a point cloud data quality automatic optimizing method and system. The point cloud data quality automatic optimizing method includes the following steps of firstly, collecting initial point cloud data for a target object to be reconstructed to obtain initial disperse point cloud; secondly, conducting primary data cleaning on the obtained initial disperse point cloud to obtain an LOP sampling model; thirdly, using a Possion surface reconstruction method on the obtained initial disperse point cloud to obtain a Possion reconstruction point cloud model; fourthly, conducting registration on the obtained Possion reconstruction point cloud model and the obtained initial disperse point cloud through the iterative closet point algorithm; fifthly, calculating the weights of surrounding points to each point on the current registered model, and comparing the weights with a threshold value to determine whether an area where each point is located needs to be scanned repeatedly or not, wherein the surrounding points are placed in the area of the obtained LOP sampling model, a position, corresponding to the point of the current registered model, of the obtained LOP sampling model serves as the circle center of the area, and a certain distance serves as the radius of the area. The invention further relates to the point cloud data quality automatic optimizing system.

Description

technical field [0001] The invention relates to a method and system for automatically optimizing the quality of point cloud data. Background technique [0002] Three-dimensional modeling is a hot problem in the field of computer science. Its modeling methods are mainly divided into two categories according to the source of data: modeling methods based on three-dimensional scattered point data and modeling methods based on images (single frame, multi-frame, sequence). The 3D laser scanning system can quickly obtain the 3D position and geometric texture information of the target object surface, and is widely used in the field of 3D modeling because of the relatively high accuracy of the data obtained. However, most of the existing modeling methods use a specific scanning instrument to scan the target object to obtain the original point cloud data required for modeling, and then use a series of artificially designed precision optimization algorithms in the later offline state ...

AI Technical Summary

Problems solved by technology

It can be seen that most of the existing high-precision 3D modeling technologies are highly dependent on the measurement accuracy of hardware equipment, and at the same time require more manual participation

Images