摘要
筒子纱染整车间纱笼坐标的检测方法是借助激光和尺子进行人工测量,存在检测效率低的问题。针对纱杆顶端圆的测量及定位这一重要环节,本文提出了一种结合连通域和霍夫(Hough)梯度的顶端圆检测方法,对预处理后的轮廓进行梯度计算,Hough圆心位于连通区域内则为有效圆。然后,通过区域面积和圆度筛选出的区域质心坐标作为圆心坐标,并将输出数据显示到人机交互界面中。经现场验证,该方法完成一个纱笼坐标的检测只需要6min,大大提高了检测效率,误差可限制在1mm以内时,准确率和稳定性能够达到95%,满足实际测量需要。
The measuring method of sarong coordinates in creel dyeing and finishing workshop is manual measurement by using laser and ruler,with the problem of low detection efficiency.For the core step of measuring and locating the yarn tip,an algorithm of detecting the circle in the tip,which combines the connected domain and the Hough gradient,is proposed.This method calculates the gradient of the preprocessed contour.The effective circle is defined as the center of Hough circle within the connected region.Afterwards,the coordinate of tip circle is the centroid coordinate of the certain contour according to the contour area and roundness.This result is shown in the human-computer interaction interface.The experiments are carried out in the real sense.The results show that only six minutes is spent on computing the coordinate of a cage,which greatly improves the detection efficiency.The measurement error is up to 1mm.The accuracy and stability can reach 95%,which meets the needs of actual measurement.
作者
冉宇辰
黄民
王文胜
李天剑
RAN Yu-chen;HUANG Min;WANG Wen-sheng;LI Tian-jian(School of Electromechanical Engineering Beijing Information Science and Technology University,Beijing 100192,China)
出处
《机械设计与制造》
北大核心
2022年第10期94-99,共6页
Machinery Design & Manufacture
基金
国家重点研发计划(2017YFB1304004)
北京市科技计划(Z191100001419009)。
关键词
机器视觉
检测
机器人控制
圆形
Machine Vision
Detection
Robot Control
Circular