一种基于多源数据融合的文物三维建模方法

A 3D Modeling Method for Artifacts Based on Multi-source Data Fusion

针对基于单一数据源对建筑物进行三维模型时,建模质量、精度、完整性具有一定的局限性使得后期修模耗时、耗力、效率低下等问题,以昆明市某文物建筑为研究对象,基于多源数据融合技术,提出一种基于计算机视觉理论中的基本原理进行多源坐标系的统一、基于主成分分析进行初始点云配准的方法,进而提出一种融合多种数据源对文物建筑进行精密三维建模的方法,并比较该方法与单一数据源建模方法的优劣。实验结果证明基于多源数据融合进行三维建模的方法,使各种古建筑三维建模技术优缺互补,规避了使用单一激光点云而无法满足古建筑顶部及局部细节三维建模的问题,使建模数据采集的手段多样化。

Based on a single data source for building 3d model,the modeling quality,accuracy,integrity,has certain limitations allow late module and time-consuming,low power consumption,efficiency and other issues,a cultural relic buildings in kunming as the research object,based on multi-source data fusion technology,puts forward the basic principle based on computer vision theory to the unity of multi-source coordinate system,based on principal component analysis method for the initial point cloud registration,and then put forward a kind of fusion of a variety of data sources for historical building precision of 3d modeling method,and compares the pros and cons of this method with single data source modeling methods.Experimental results show that based on multi-source data fusion method for 3d modeling,make each modeling technology to carry on the complementary advantages,solves the oblique photography dioramas details of complementary and precise modeling,laser point cloud can't satisfy the top and detail modeling problem,enrich the modeling means of data collection,meet the needs of the project for all kinds of data accuracy.