层次化的文物碎片点云模型匹配方法  

Hierarchical Matching Method for Point Cloud Model of Cultural Relics Fragments

作　　者：汤慧 赵夫群 耿国华[2] TANG Hui;ZHAO Fu-qun;GENG Guo-hua(School of Information,Xi'an University of Finance and Economics,Xi'an Shaanxi,710100,China;School of Information Science and Technology,Northwest University,Xi'an Shanxi,710127,China)

机构地区：[1]西安财经大学信息学院,陕西西安710100 [2]西北大学信息科学与技术学院,陕西西安710127

出　　处：《计算机仿真》2024年第11期239-243,共5页Computer Simulation

基　　金：国家自然科学基金(61731015);全国统计科学研究项目(2021LY057);陕西省自然科学基础研究计划项目(2021-JQ-765)。

摘　　要：碎片匹配是文物虚拟复原的关键步骤之一,匹配精度将直接影响到文物虚拟复原的结果。针对文物碎片点云数据模型具有数据量大、分辨率差异大等特点,提出一种层次化的碎片匹配方法。首先,采用GH-LS3D(least-squares 3D)算法对文物碎片进行粗匹配,通过使用非线性Gauss-Helmert模型来确定待匹配碎片的刚体变换参数;然后,采用一种基于尺度因子和模拟退火因子的改进ICP(iterative closest point)算法,实现碎片精确匹配;最后,采用兵马俑碎片点云来验证该匹配方法。实验结果表明,上述层次化的碎片匹配方法的匹配精度和匹配速度比已有算法至少可以分别提高约10%和15%,是一种有效的文物碎片匹配方法。Fragment matching is one of the key steps of cultural relics virtual restoration,the matching accuracy will directly affect the results of cultural relics virtual restoration.Aiming at the characteristics of large amount of data and large resolution difference of Cultural Relic Fragment point cloud data model,a hierarchical fragment matching method was proposed.Firstly,GH-LS3D(least-squares 3D)algorithm was used to roughly match the cultural relics fragments,and the rigid body transformation parameters of the fragments to be matched were determined by using the nonlinear Gauss Helmert model;Then,an improved ICP(iterative closest point)algorithm based on scale factor and simulated annealing factor was used to realize accurate fragment matching;Finally,the matching method was verified by using Terracotta Warriors fragment point clouds.The experimental results show that the matching accuracy and speed of the hierarchical fragment matching method could be improved by at least 10%and 15%respectively compared with the existing algorithms.It is an effective method for matching cultural relics fragments.

关 键 词：文物复原 碎片匹配 模拟退火 尺度因子 迭代最近点 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]