摘要
提出一种平滑度欧式聚类点云分割算法,用于实现对Kinect点云的快速、准确分割。首先介绍了Kinect点云的采集和滤波方法,然后在传统欧式聚类算法基础上提出了一种平滑度欧式聚类分割算法,通过加入平滑阈值的约束来防止过度分割或分割不足的问题,并保持了较快的分割速度。通过对工业机器人获取的阀门点云数据进行实验,证明了算法的有效性。
An Euclidean cluster extraction algorithm with the smoothness parameter for Kinect point cloud is presented to achieve fast,accurate segmentation.First,the Kinect point cloud collection and filtering method is introduced,then an Euclidean cluster extraction algorithm with the smoothness is proposed by adding a smoothness threshold to traditional Euclidean clustering algorithm,which can prevent the problem of over-segmentation or inadequate split,and maintain a rapid segmentation speed.The experiments of valve point cloud data acquired by an industrial robot prove the effectiveness of the algorithm.
出处
《测控技术》
CSCD
2016年第3期36-38,共3页
Measurement & Control Technology
基金
中央高校基本科研业务费专项资金青年教师资助计划项目(ZY20140211)
关键词
分割
点云数据
欧式聚类
平滑度
segmentation
point cloud
Euclidean clusters
smoothness
