点云
稳健性(进化)
焊接
特征提取
计算机科学
聚类分析
点(几何)
曲线拟合
算法
人工智能
计算机视觉
工程类
几何学
数学
机械工程
生物化学
化学
机器学习
基因
作者
Shuai Yang,Xiaorui Shi,Xincheng Tian,Yan Liu
标识
DOI:10.1109/ddcls55054.2022.9858360
摘要
In this paper, a method based on point cloud for weld seam extraction of intersecting pipes workpiece is proposed. Firstly, a lightweight clustering algorithm based on the normal vector direction was proposed to extract the feature points of the weld seam. After filtering the feature points, a novel key points extraction algorithm was proposed to extract the key points of the weld seam. After that, a non-Uniform Rational B-Spline curve was used to approximate the key points and the preliminary spatial weld curve was obtained. Finally, an iterative algorithm was adopted to optimize the generated weld curve. The proposed method is verified on the ideal and real intersecting pipes point cloud models. Experiments show that the proposed method has high accuracy and robustness, and can accurately extract the intersecting pipes weld seam curve, providing guidance for the subsequent accurate automatic welding.
科研通智能强力驱动
Strongly Powered by AbleSci AI