期刊导航

论文摘要

基于方向角的散乱点云三角剖分算法

One unorganized points cloud triangulation algorithm based on orientation angle

作者:魏永超(四川大学电子信息学院光电科学技术系);苏显渝(四川大学 电子信息学院,四川 成都 610064)

Author:Wei Yong-Chao();苏显渝()

收稿日期:2008-07-01          年卷(期)页码:2009,41(4):202-207

期刊名称:工程科学与技术

Journal Name:Advanced Engineering Sciences

关键字:曲面重建,法线矢量,K邻域,三角化,方向角

Key words:surface reconstruction, normal vector, k-nearest neighbors, triangulation, orientation angle

基金项目:国家自然科学基金

中文摘要

针对直接三角剖分困难,分片三角剖分需要人工干预,且算法效率都很低下问题,提出了高效自动的在特征基点根据方向角进行分片投影三角剖分。算法的主要步骤分为两步:首先从只有位置信息的点云中提取出邻域、矢量和形状索引信息,并利用形状索引信息得到特征基点;然后以特征基点为参考点根据方向角对点云进行分片,每个片进行特征基点切平面投影三角剖分,通过三角剖分有效性处理,连接成最终的网格曲面。实验结果表明算法可以自动高效的重叠和非重叠散乱点云,且可以有效避免曲面自交。

英文摘要

One novel slicing triangulation algorithm was proposed, which dose not need manual intervention and can triangulate both overlapping and non-overlapping points cloud highly efficiently. The algorithm contains two main stages: points cloud preprocessing and triangulation. In preprocessing phase, the information of KNN、vector and shape index are extracted from points’ location information; In triangulation stage, if one new basis point is picked out, one new patch will be created based on the orientation angle between the basis point and each other unprocessed point, then the patch is projected onto the basis point’s tangent plane and triangulated, after verifying triangles in 3D space, the triangulation results of the patch are saved. Final triangulation results are output until there is no new basis point available. The experiment results confirm the proposed algorithm is effective and highly-efficient.

关闭

Copyright © 2020四川大学期刊社 版权所有.

地址:成都市一环路南一段24号

邮编:610065