首页|基于双判别准则的散乱点云孔洞识别算法

基于双判别准则的散乱点云孔洞识别算法

扫码查看
针对传统点云孔洞识别算法效率不高且易受尖锐点影响等问题,提出一种半圆盘准则与最大夹角准则相结合,并引入特征值约束的孔洞识别算法.首先建立kdtree索引搜寻邻域点,利用半圆盘准则提取准边界点;然后利用最大夹角准则去掉冗余点,进而精确提取孔洞边界点,再基于最小特征值对边界点进行筛选,去除误识别的尖锐点;最后对多个孔洞点集进行聚类,完成孔洞识别.实验结果表明,该算法能够消除尖锐点的影响,准确、高效地识别孔洞边界点,为后续孔洞修补操作打下基础.
Hole recognition algorithm based on double discriminant criterion in scattered point cloud
Aiming at the problem that the traditional point cloud hole identification algorithm is inefficient and easily affected by sharp points,a hole identification algorithm that combines the semicircle criterion and the maximum angle criterion and introduces eigenvalue constraints is proposed.First,a kdtree index is established to search for neighborhood points,and the semi-disk criterion is used to extract quasi-boundary points.Then the maximum angle criterion is used to remove redundant points to accurately extract hole boundary points.Then the boundary points are screened based on the minimum eigenvalue to eliminate misidentifications.sharp points,and finally cluster multiple hole point sets to complete hole identification.Experimental results show that this algorithm can eliminate the influence of sharp points,identify hole boundary points accurately and efficiently,and lay the foundation for subsequent hole repair operations.

scattered point cloudkdtreethe halfdisc criterionboundary extraction

孙启翔、郭敏、吕源治、李贞兰

展开 >

长春工业大学计算机科学与工程学院,吉林长春 130102

中国科学院长春光学精密机械与物理研究所,吉林长春 130033

吉林大学白求恩医学部,吉林长春 130012

散乱点云 kdtree 半圆盘准则 边界提取

2024

长春工业大学学报
长春工业大学

长春工业大学学报

影响因子:0.282
ISSN:1674-1374
年,卷(期):2024.45(5)