摘要
为解决汽车线束包胶机胶带人工目检效率及准确率低的问题,基于机器视觉技术提出了一种针对汽车线束包胶胶带切断与入槽状态的自动化视觉检测方法。首先获取胶带图像并灰度处理,然后基于归一化互相关(NCC)的图像匹配方法精确定位并提取胶带槽口检测区域,最后基于支持向量机(SVM)算法识别判断胶带切断与入槽情况。试验结果表明,该自动化视觉检测方法显著提高了胶带的检测效率及准确率。
In order to solve the problem of low efficiency and low accuracy in manual eye detection of adhesive tapes on automobile harness wrapping machine,an automatic visual inspection method for the cutting and grooving state of adhesive tapes on automobile harness wrapping is proposed based on machine vision technology.Firstly,the image of adhesive tape is obtained and processed in gray scale,and then the detection area of tape notch is accurately located and extracted based on the image matching method of normalized cross-correlation(NCC).Finally,the cutting and grooving states of adhesive tapes are recognized and judged based on the support vector machine(SVM)algorithm.The test results show that the automatic visual inspection method significantly improves the efficiency and accuracy of tape detection.
作者
原敏乔
Yuan Minqiao(Changchun Institute of Optics,Fine Mechanics and Physics,Chinese Academy of Sciences,Changchun 130033,Jilin,China)
出处
《工程与试验》
2022年第2期33-36,136,共5页
Engineering and Test
关键词
视觉检测
胶带检测
归一化互相关
支持向量机
visual inspection
tape inspection
normalized cross-correlation
support vector machine
