摘要
针对现有的工业上的标签表面质量检测过程中存在的速度慢,精度低的问题,提出一种基于Halcon的工业标签表面的印刷图案的缺陷检测。其方法主要思想为差分思想即根据不同工业标签表面图案的区域特征进行Blob分析来定位,通过基于形状的模板匹配算法来快速查找目标区域,然后利用灰度值差影匹配算法进行缺陷检测。最后通过图像配准的方法检测缺陷的特征值,该检测方法得到的检测结果与实际存在的缺陷基本一致,而且大大提高了检测的速度和精度,达到了生产线上的质量要求。
Aiming at the problems of slow speed and low precision in the label surface quality inspection process of the existing industry,a defect detection method based on the printed pattern of the industrial label surface of Halcon is proposed.The main idea of the method is that the differential idea is to perform Blob analysis according to the regional features of different industrial label surface patterns,and the target region is quickly searched by the shape-based template matching algorithm,and then the gray value difference matching algorithm is used for defect detection.Finally,the image registration method is used to detect the feature value of the defect.The detection result obtained by the detection method is basically consistent with the actual defect,and the detection speed and precision are greatly improved,and the quality requirement on the production line is achieved.
作者
熊继淙
邢旭朋
马军
XIONG Jicong;XING Xupeng;MA Jun
出处
《科技视界》
2020年第3期36-37,共2页
Science & Technology Vision
关键词
机器视觉
标签缺陷检测
差分思想
模板匹配
Machine vision
Label defect detection
Differential thinking
Template matching