基于视觉的胶水偏位检测
|
摘要
现有热转印胶片胶水偏位人工检测方法存在效率低,由此提出一种基于机器视觉的检测方式。设计机器视觉算法,利用边缘检测算法与图像仿射变换等步骤获取只包含热转印胶片的灰度图;然后利用改进的K-means应用方法,分割出胶水层的二值图。通过本文特定的扫描方式分离出四个不同方向的胶水层部分,并计算出最终的胶水宽度,并以此作为判断依据是否存在胶水偏位问题。实验表明,本文的方法具有较好的检测效果。
The existing manual detection method for glue misalignment of thermal transfer film has low efficiency,so this paper proposes a detection method based on machine vision. Firstly,a machine vision system is designed,which uses edge detection algorithm, image affine transformation and other steps to obtain the gray image containing only thermal transfer film. Then the improved K-means application method is used to segment the binary image of the glue layer. Through the specific scanning method in this paper,four glue layer parts in different directions are separated,and the final glue width is calculated,which is used as a basis to judge whether there is glue misalignment defect. Experiments show that the method in this paper has good detection effect.
The existing manual detection method for glue misalignment of thermal transfer film has low efficiency,so this paper proposes a detection method based on machine vision. Firstly,a machine vision system is designed,which uses edge detection algorithm, image affine transformation and other steps to obtain the gray image containing only thermal transfer film. Then the improved K-means application method is used to segment the binary image of the glue layer. Through the specific scanning method in this paper,four glue layer parts in different directions are separated,and the final glue width is calculated,which is used as a basis to judge whether there is glue misalignment defect. Experiments show that the method in this paper has good detection effect.
引文
[1]胡建.热转印的分类与应用[J].丝网印刷,2014(04):19-22.
[2]李宪臣.热转印领域最前沿工艺及材料[J].丝网印刷,2013(07):24-25.
[3]胥磊.机器视觉技术的发展现状与展望[J].设备管理与维修,2016(09):7-9.
[4]周显恩,王耀南,朱青,等.基于机器视觉的瓶口缺陷检测方法研究[J].电子测量与仪器学报,2016,30(05):702-713.
[5]王宪保,李洁,姚明海,等.基于深度学习的太阳能电池片表面缺陷检测方法[J].模式识别与人工智能,2014,27(06):517-523.
[6] Dong Y W,Wan L,Shi Z M,et al.An Image Registration Algorithm Based on Improved SIFT Feature[J].Applied Mechanics and Materials,2013,347-350(347-350):3411-3415.
[7]张志斌,赵帅领,罗锡文,等.基于SURF算法的绿色作物特征提取与图像匹配方法[J].农业工程学报,2015,31(14):172-178.
[8] Otsu N.A Threshold Selection Method from Gray-Level Histograms[J].IEEE Transactions on Systems,Man,and Cybernetics,1979,9(1):62-66.
[2]李宪臣.热转印领域最前沿工艺及材料[J].丝网印刷,2013(07):24-25.
[3]胥磊.机器视觉技术的发展现状与展望[J].设备管理与维修,2016(09):7-9.
[4]周显恩,王耀南,朱青,等.基于机器视觉的瓶口缺陷检测方法研究[J].电子测量与仪器学报,2016,30(05):702-713.
[5]王宪保,李洁,姚明海,等.基于深度学习的太阳能电池片表面缺陷检测方法[J].模式识别与人工智能,2014,27(06):517-523.
[6] Dong Y W,Wan L,Shi Z M,et al.An Image Registration Algorithm Based on Improved SIFT Feature[J].Applied Mechanics and Materials,2013,347-350(347-350):3411-3415.
[7]张志斌,赵帅领,罗锡文,等.基于SURF算法的绿色作物特征提取与图像匹配方法[J].农业工程学报,2015,31(14):172-178.
[8] Otsu N.A Threshold Selection Method from Gray-Level Histograms[J].IEEE Transactions on Systems,Man,and Cybernetics,1979,9(1):62-66.