摘要
提出了一种基于机器视觉的内螺纹检测算法,并在此基础上设计了一套多螺纹孔检测系统。首先从CCD获取的图像中提取出ROI(region of interest)区域,然后利用中值滤波去除图像表面噪声,通过形态学开闭运算对所得图像进行处理,最后计算螺纹图像沿螺孔母线方向的灰度曲线极值得到螺纹牙数,进而对于螺孔合格与否进行判断。实验结果表明,基于该算法的多螺孔检测系统能够有效地提高螺孔的检测速度,具有良好的检测精度,对外界光照变化具有良好的鲁棒性。
An internal thread detection algorithm based on machine vision is proposed,and a set of multi-threaded hole detection system is designed on this basis.This method first extracts the ROI(region of interest)area from the image acquired by the CCD,then uses median filtering to remove image surface noise,and processes the resulting image through morphological opening and closing operations,and finally calculates the direction of the thread image along the screw hole generatrix The extreme value of the gray-scale curve obtained the number of thread teeth,and then judge whether the threaded hole is qualified or not.Experimental results show that the multi-threaded detection system based on this algorithm can effectively improve the detection speed of threaded holes,has good detection accuracy,and has good robustness to external light changes.
作者
肖庆阳
方建儒
张效民
刘冬
丛明
XIAO Qing-yang;FANG Jian-ru;ZHANG Xiao-min;LIU Dong;CONG Ming(Dalian Yaming Automotive Parts Co.,Ltd.,Dalian Liaoning 116041,China;School of Mechanical Engineering,Dalian University of Technology,Dalian Liaoning 116024,China)
出处
《组合机床与自动化加工技术》
北大核心
2021年第1期126-127,共2页
Modular Machine Tool & Automatic Manufacturing Technique
关键词
螺纹孔检测
形态学
特征识别
threaded hole detection
morphology
feature recognition
