摘要
针对目前螺纹检测精度较低,在线检测实现较难的问题,提出了一种基于机器视觉的螺纹缺陷检测方法.通过工业CMOS相机在线采集螺纹零件图像,经图像预处理、二值化后,进行螺纹的边缘提取及缺陷检测等.该算法由Matlab编程实现.实验结果表明,该螺纹缺陷检测方法能有效识别无螺纹、少螺纹、螺距不规则等外观缺陷,且一个螺纹的检测时间在300ms以内,满足检测要求.
A method for detecting thread defect based on machine vision is aiming at the problem of low precision and realizability.Thread images are taken by CMOS camera;and extraction of thread edge and detection of thread defect can be done after preprocessing and binarization.The method is realized by Matlab programming.Experimental results show that the method can identify non-thread,less-thread,uneven pitch and other defects.The detection time of a thread is less than 300 ms that meets the detection requirements.
出处
《三峡大学学报（自然科学版）》
CAS
2015年第2期64-66,共3页
Journal of China Three Gorges University：Natural Sciences
基金
国家自然科学基金(51407104)
关键词
机器视觉
螺纹
边缘提取
缺陷检测
machine vision
thread
extraction of thread edge
detection of thread defects
