摘要 针对环形器件表面检测中存在的问题,文中以环形金属涂层陶瓷为研究对象,提出了一种基于机器视觉的表面缺陷检测方法。该方法使用Canny边缘检测和Hough变换快速定位同心圆轮廓,进而提取金属涂层区域。为了将极坐标转换为直角坐标,文中采用径向投影方法将环形金属涂层图转换为矩形状图。最后,对连通域面积和周长进行分析,确定环形陶瓷金属涂层是否存在缺陷。实验结果表明,对于缺损面积大于0.5 mm×0.5 mm的样品,该算法的检测正确率为98%。 In view of the problems in the surface inspection of annular devices,the current study takes the annular metal-coated ceramics as the research object and proposes a surface inspection method based on machine vision.The proposed method uses Canny edge detection and Hough transform to quickly locate concentric circles,and then extracts metal-coated areas.In order to convert polar coordinates to rectangular coordinates,the radial projection method is used to convert the ring-shaped metal coating map to a rectangular shape.Finally,the area and perimeter of the connected domain are analyzed to determine whether the circular ceramic metal coating has defects.The experimental results show that the detection accuracy of the algorithm is 98%for samples with a defect area greater than 0.5 mm×0.5 mm.
出处 《电子科技》 2021年第6期28-33,共6页 Electronic Science and Technology
基金 国家科技支撑计划(2014BAH28F04) 湖南省科学研究项目(15K009,17K004) 湖南省重点研发计划(2018GK2054)。
关键词 环形器件表面 机器视觉 边缘检测 HOUGH变换 缺陷检测 径向投影 坐标转换 检测系统 ring device surface machine vision edge detection Hough transform defect detection radial projection coordinate transformation detection systems
