基于误差阈值控制的ICP点云配准算法研究

Research on Point Cloud Registration Algorithm of ICP based on Control with Error Threshold in Three-Dimensional Scanning of Cultural Relics

文物在保存、挖掘、运输过程中,难免会受到自然及人为因素的破坏,利用计算机技术对具有艺术价值、人文价值和科学研究价值的文物进行扫描拼接、修复和重建,还原文物形状,一直是考古界重点关注的热点。文章采用三维激光扫描获得物体表面的数字化点云信息,并通过ICP算法进行点云配准,在计算机中生成三维数据模型。在点云配准过程中对常用的点云配准算法进行了实验对照,结果发现基于误差阈值控制的ICP点云配准算法相比于传统的ICP算法更具稳定性和收敛性,更能节约运算资源。本算法为学术研究、艺术鉴赏、信息分享与文化传承及利用提供了更多、更精确的文物数字化方法。

During the excavation of cultural relics,it is inevitable that they will be destroyed,by natural and human factors.Using computer technology to digitally scan,splice,repair and virtually reconstruct these cultural relics with artistic value,humanistic value and scientific research value,and restore cultural relics,has always been the focus of archaeological circles.In this paper,the key technology of 3D virtual reconstruction,point cloud registration,is studied.Digital information is obtained by 3D scanning,and point cloud registration is carried out,which provides more and more accurate digital methods for academic research,art appreciation,information sharing and cultural inheritance and utilization.