摘要
为了提高四通阀焊接的速度和精度,提出了四通阀自动钎焊定位技术及系统。运用基于图像处理方法的计算机视觉技术,实现了对空调零件四通阀钎焊坐标的快速准确定位。采用近红外背光源进行照明,实现了对四通阀圆心坐标及侧面高度的自动检测。采用近红外背光源能使成像边缘更清晰易辨,具有环保节能、寿命长等优点。该系统主要通过阈值分割和PCA算法分别实现对四通阀铜柱圆轮廓提取和铜柱高度的水平投影,通过形状匹配和直线拟合获得圆心相对坐标和高度,实现对四通阀钎焊定位的检测。该系统具有速度快、准确率高、系统运行稳定等优点,在工业上的四通阀钎焊中具有较好的应用性和广阔的发展前景。
To improve the speed and precision of four-way valve brazing,an automatic brazing inspention system was proposed.Based on machine vision the four-way valve brazing coordinate was located accurately.Using near-infrared backlight source for lighting the automatic dectections of fourway valve center and side heigh were realize.Using near-infrared backlight the image edge might make more clear and easy to argue,with the advantages of environmental protection and energy saving,long life,et al.The system might achieve the level projection of four-way valve circle contour extraction and column height mainly through threshold segmentation and PCA algorithm respectively.Through shape matching and linear fitting the center of circle relative coordinate and height were obtained to achieve the four-way valve brazing inspectiong system.The four-way valve brazing inspection has the advantages of high speed,high accuracy,stable system operation and so on.It has good applications and broad development prospects in industries.
作者
孔明
贺楚红
王英军
黄玮
Kong Ming He Chuhong Wang Yingjun Huang Wei(China Jiliang University, Hangzhou, 310018)
出处
《中国机械工程》
EI
CAS
CSCD
北大核心
2016年第22期3043-3047,共5页
China Mechanical Engineering
基金
国家自然科学基金资助项目(51375467)
关键词
图像处理
图像分析
灰度直方图
测量
image processing
image analysis
gray histogram
measurement