摘要
计算机处理速度的提高和深度相机的广泛应用使点云数据在工业领域中的应用越来越广泛。针对巡检机器人非固定场景中小孔定位问题,提出小孔边界检测提取算法。对点云进行预处理;求取目标点对应近邻点的切平面上的投影点,通过投影点算出质心点位置,由质心位置将圆盘分为两个半圆盘,通过两个半圆盘中投影点数量比值作为边界点判定条件。实验表明,该算法可以较好地提取出边界点使得噪声对结果的影响成比例减小,鲁棒性好,运行速度快且稳定。
The improvement of computer processing speed and the wide application of depth camera make the point cloud data more and more widely used in the industrial field.To address the problem of locating small and medium holes in non fixed scene of patrol robot,a small hole boundary detection algorithm is proposed.The point cloud is preprocessed.The projection point on the tangent plane of the target point corresponding to the adjacent point is obtained.The center of gravity point is calculated by the projection point.The disc is divided into two half disks by the center of gravity position.The ratio of projection points in two semi disks is used as the boundary point judgment condition.Experimental results show that the algorithm can well extract boundary points,reduce the influence of noise on the result by times,has good robustness,and gain fast and stable operation speed.
作者
王宪伦
丁文壮
孙旭祥
WANG Xianlun;DING Wenzhuang;SUN Xuxiang(College of Electromechanical Engineering,Qingdao University of Science and Technology,Qingdao 266100,China)
出处
《机械制造与自动化》
2021年第5期39-41,52,共4页
Machine Building & Automation
关键词
点云
孔洞边界检测
半圆盘度量准则
场景定位
point cloud
hole boundary detection
half-disc criterion
scene positioning
作者简介
第一作者:王宪伦(1978—),男,山东济宁人,博士,主要从事机器人方向的研究。
